WO2013040445A1 - Arrays for cell-based screening and uses thereof - Google Patents

Abstract

In some aspects, the invention provides arrays of use for cell-based screening.

Description

ARRAYS FOR CELL-BASED SCREENING AND USES THEREOF

Related Applications

[0001] This application claims priority to and the benefit of U.S. Provisional Application No. 61/535,267, filed on September 15, 2011. The entire teachings of the above application(s) are incorporated herein by reference.

Background of the Invention

[0002] Gain- and loss-of-function genomic screens have the potential to reveal the mechanistic underpinnings of a wide range of cell-based biological processes and already have been used successfully to gain insights into phenomena as diverse as host-pathogen interactions, mitotic regulation, and synthetic lethality. However, functional screens have important limitations that have hindered their widespread deployment. For example, although arrayed screens in microtiter plates can provide discrete measurements of proliferation-, reporter-, and image-based phenotypes in individual populations of cells, they can be restricted by factors such as high cost per well, limited throughput, and the requirement for specialized screening facilities equipped with appropriate fluid handling equipment. Pooled screening approaches can reduce costs and allow extended assay time courses but have a number of limitations. For example, such approaches can require complex deconvolution and use of assays that permit the selection of hits or candidates from a mixed cell population. There is a significant need for technologies that would facilitate high throughput functional genomic screening.

Summary of the Invention

[0003] The present invention relates at least in part to arrays for use in cell-based screening. In some aspects, the invention provides an array comprising a surface having multiple discrete features, wherein each feature comprises a cell adhesive material and agent, and wherein the features are separated by regions that inhibit eukaryotic cell migration. In some embodiments the agents are reversibly associated with the cell adhesive material. In some embodiments the regions that inhibit mammalian cell migration inhibit eukaryotic cell adhesion. In some embodiments the agents comprise nucleic acids, polypeptides, or small ] molecules. In some embodiments the agents comprise viruses. In some embodiments the viruses are lentiviruses. In some aspects, an array is seeded with eukaryotic cells, e.g. mammalian cells. In some aspects, cells are substantially confined to features comprising a cell adhesive material. In some embodiments, agents are released from the features, enter overlying cells, and result in transcription of open reading frame or RNAi agent within the cells.

[0004] In some aspects, the invention provides a method of making an array for infection or transfection of eukaryotic cells, the method comprising: (a) providing a surface comprising spots that comprise a cell adhesive material, wherein the spots are separated by a cell adhesion resistant material; and (b) depositing an agent on top of each spot. In some embodiments the agents comprise nucleic acids, polypeptides, or small molecules. In some embodiments the agents comprise viruses. In some embodiments the viruses are lentiviruses.In some aspects, the invention provides methods of making any of the inventive arrays. In some aspects, the invention provides methods of using any of the inventive arrays, e.g., to perform any of a wide variety of cell-based screens.In some aspects, a cell-seeded array is contacted with a test compound, and the effect of the test compound on cell phenotype is assessed.

[0005] In some aspects, the invention provides methods of predicting tumor or tumor cell resistance to a MAPK pathway inhibitor.

[0006] Certain techniques of cell biology, cell culture, molecular biology, microbiology, recombinant nucleic acid (e.g., DNA) technology, immunology, etc., which are within the skill of those of ordinary skill in the art, may be of use in aspects of the invention. Non- limiting descriptions of certain of these techniques are found in the following publications: Ausubel, F., et al., (eds.), Current Protocols in Molecular Biology, Current Protocols in Immunology, Current Protocols in Protein Science, and Current Protocols in Cell Biology, all John Wiley & Sons, N.Y., editions as of 2008; Sambrook, Russell, and Sambrook, Molecular Cloning: A Laboratory Manual, 3^rd ed., Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 2001 ; Harlow, E. and Lane, D., Antibodies - A Laboratory Manual, Cold Spring Harbor Laboratory Press, Cold Spring Harbor, 1988; Freshney, R.I., Culture of Animal Cells: A Manual of Basic Technique, Wiley-Liss, 5th edition, 2005, or 6^th edition, published online March 9, 2011; Burns, R., Immunochemical Protocols (Methods in Molecular Biology) Humana Press; 3rd ed., 2005. All patents, patent applications, websites, databases, scientific articles, and other publications mentioned herein are incorporated herein by reference in their entirety. In the event of a conflict or inconsistency with the specification, the specification shall control. The Applicants reserve the right to amend the specification based on any of the incorporated references and/or to correct obvious errors. None of the content of the incorporated references shall limit the invention.

]

Brief Description of the Drawing

[0007] Figure 1. The MicroSCALE screening platform, (a) General schematic depicting screening with MicroSCALE. (b) MieroSCALEs printed with GFP-expressing lentiviruses were seeded with the indicated cell lines and fixed/imaged after 4 d (FITC channel shown). Scale bar, 1 mm. (c) MieroSCALEs printed with GFP-expressing lentiviruses were seeded with U2-OS cells and incubated for 4 d with (right) or without (left) puromycin selection during days 1-4 (blue: Hoechst; green: FITC/GFP). Scale bar, 1 mm. 10008] Figure 2. MicroSCALE is compatible with diverse screening applications, (a) MicroSCALE slides were seeded with U2-OS cells, selected with puromycin beginning on day 1, grown for the indicated time intervals, and then fixed/stained (Hoechst). Cell numbers were determined by counting the number of nuclei per spot, (b) MieroSCALEs were seeded with the indicated cell lines, selected with puromycin beginning on day 1 , and fixed/imaged (Hoechst) on day 6. Relative viabilities were determined by quantifying the number of cells per feature, (c) MieroSCALEs were seeded with U2-OS cells, selected with puromycin, and fixed/imaged (Hoechst, P-S6) on day 5, Phospho-S6 levels were determined by measuring the total P-S6 staining intensity (background subtracted and normalized to Hoechst intensity), (d) The constructs in the CCSB/Broad Kinase ORF collection were printed in duplicate or triplicate to form 25 ^χ 75 mm, 1632-feature Kinome ORF MieroSCALEs. Features are 600 μηι in diameter with 750 μηι center-to-center spacing. The two features in the bottom right hand corner of each 6x6 sub-grid arc control spots containing no virus. Arrays were seeded with U2-OS cells, selected with blasticidin, and fixed/stained/imaged after 6 d (Syto82). Scale bars, 10 mm and 1 mm (inset), (e) Results of a screen in 5-&<l '^'< '^w'^;-mutant melanoma cells (A375) to identify ORFs whose overexpression decreases sensitivity to PLX4720 (1 μΜ). Average viability scores (drug/vehicle) of individual ORFs are shown, with the top 10% of ORFs shaded in gray. Red bars represent the top hits from an analogous multiwell plate-based screen and blue arrows indicate mutant positive control ORFs. (f) Images of replicate spots on a PLX4720-treated array, including mutant hits (blue), wild-type ORF hits (red), and controls (black).

[0009] Figure 3. Integrated MicroSCALE functional screens and pharmacogenomic analyses reveal genetic modifiers to targeted inhibitors in melanoma, (a) Schematic depicting an integrated screening approach to discover high priority resistance genes and pathways, (b) Heat map showing the results of modifier screens. Columns represent individual arrays (2-5 replicate arrays were screened for each drug) and rows represent the average proliferation score of each ORF (drug/vehicle). Unsupervised hierarchical clustering of rows and columns was performed (for simplicity, the dendrogram representing the results of row-based clustering is not shown). Scale bar indicates Z-scores (standard deviations from the column mean; see Methods), (c) Hits that decrease the sensitivity of A375 cells to MAPK pathway inhibitors and were validated in secondary assays fall into defined functional categories, (d) Top: Heat map depicting PLX4720 and AZD-6244 GI50 values for a panel of 25 B-RAF^V600E -mxAmt melanoma cell lines. GI50 values are row (drug) normalized and Z- transformed. Bottom: Heat map showing single-sample GSEA scores (see Methods) for three gene sets annotating NF-κΒ activation. The matching scores (see Methods) of the NF- KB gene sets against PLX4720's GI50 profile reveal a highly significant enrichment of those gene sets in PLX4720- and AZD-6244-resistant cell lines. The histogram depicts the matching scores of 3,264 gene sets (MSigDB/C2 v3.0) against PLX4720's GI50 profile, with the scores of the NF-κΒ gene sets highlighted by green lines.

[0010] Figure 4: NF-κΒ activation is associated with resistance to MAPK pathway inhibitors in B-RAF^V6ME melanomas, (a) NF-κΒ pathway activation by IKBKB or TRAF2 overexpression, or by the addition of exogenous TNFa (25 ng/mL), increases the GI50 concentration for RAF, MEK1/2, and ERK inhibitors in four B-RAF^vmE melanoma cell lines relative to a control (MEK1 overexpression). (b) In B-RAF^1,60011 melanoma patient-derived short-term cultures, gene expression signatures of NF-κΒ activation accurately predict resistance to MAPK pathway inhibitors.

[0011] Supplementary Figure 1: Strategy for restricting cell adhesion, spreading, and infection to only the printed features on the slide surface, (a) GFPexpressing lentiviruses printed on aminosilane-coated glass slides following concentration yield high levels of infection but varying localization (GFP expression is shown; note that cells adhere uniformly to the entire surface of aminosilane coated slides). Scale bar, 1 mm. (b) Schematic depicting the strategy for localizing cell adhesion, spreading, and infection, (c)

Polyacrylamide hydrogel-coated glass slides printed in two stages - first with gelatin.second with lentiviruses - exhibit cell adhesion and infection only on printed features (GFP expression shown; note that cells adhere only to the printed features). Scale bar, 1 mm. Cells were incubated with cells for 5 days prior to imaging in panes (a) and (c).

[0012] Supplementary Figure 2: Lentiviruses are purified of serum proteins and concentrated prior to printing in order to efficiently bind the microarray surface and infect adjacent cells, (a) Lenti viral p24 staining of array features printed using concentrated lentiviruses (5 x 108 IFU/mL) suspended in serum-free or 10% serumcontaining print buffer prior to printing. Scale bar, 500 μΜ. (b) GFP expression on microarray spots printed on amino silane-coated glass slides using lentiviruses at the indicated titers in serum-free print buffer and seeded with the indicated cell lines. Images were obtained 5 d post-seeding. Scale bar, 500 μΜ.

[0013] Supplementary Figure 3: Low-speed, high-throughput lentivirus concentration using polyelectrolyte complexation yields concentrated viruses suitable for microarray printing, (a) Schematic depicting the lentivirus concentration method, (b) Titer (IFU/mL) of lentiviruses before concentration (Uncon) or following polyelectrolyte complexation, centrifugation for indicated times (at 1150g), aspiration to remove residual media, and resuspension in a smaller volume of printing buffer. Indicated viral titers following 10, 20, and 60 minute centrifugation represent a recovery of approximately 80% of the virus in the original supernatant.

[0014] Supplementary Figure 4: Visualization of puromycin selection of infected cells on MicroSCALEs. Polyacrylamide hydrogel-coated slides were printed with features containing gelatin and either GFP-expressing lentiviruses encoding the puromycin resistance gene (PAC) or no virus. Arrays were seeded with A549 cells and treated with puromycin (or vehicle) on day 1 post-seeding. Images of individual features following 1, 3, and 4 days of puromycin selection demonstrate the selection of virus-infected cells on (+puro/+virus) features. Scale bar, 250 μΜ.

[0015] Supplementary Figure 5: High-throughput quantification of cell number on MicroSCALE features using a DNA microarray scanner, (a) Total intensity of the Syto82 DNA stain on individual spots correlates linearly with the number of cells present on those spots, (b) Growth curves generated by counting the number of cells on individual MicroSCALE spots (Fig 2A) can be rapidly reproduced by quantifying the total Syto82 DNA staining intensity on each spot using a DNA microarray scanner.

[0016] Supplementary Figure 6: The CCSB/Broad Kinase ORF collection is compatible with MicroSCALE-based screening, (a) Lentiviral ORFs representing the range of sizes found in the kinase collection were produced, concentrated, and printed as microarrays. Arrays were seeded with HBL- 100 cells and blasticidin selected for 4 days. Blue: Hoeschst, Red: Phalloidin. (b) Schematic of the pLX-BLAST-V5 lentiviral expression vector used for various ORF-screens and subsequent validation studies, (c) Description of the collection. [0017] Supplementary Figure 7: Schematic outlining the MicroSCALE OKI' -based functional screening pipeline used to uncover kinases that drive resistance to targeted therapies in B-RAFV600E mutant melanoma cells.

[0018] Supplementary Figure 8: Sensitivity of A375 (BRAFV600E mutant melanoma) and a second reference cell line, MDA-MB-453 (Her2 amplified, PIK3CA mutant breast cancer), to a RAF inhibitor (ex: PLX4720), a MEK1/2 inhibitor (ex: AZD-6244), an mTOR inhibitor (ex: Torin-1), and an mTOR/PI3 inhibitor (ex: BEZ- 235). Cells were seeded in 96-well plates, treated with drugs after 1 d, and assayed for viability after 4 d. Error bars represent one standard deviation.

[0019] Supplementary Figure 9: Confirmation of hits identified in MicroSCALE primary screens, (a) Approximately 30-40 ORFs scoring in each primary screen were validated in secondary assays in 96-well plate format. Confirmed hits are those yielding viability scores (viability in the presence of drug / viability in the presence of vehicle) greater than MEKl, a neutral control, (b) Venn diagram representing the overlap in ORFs scoring in RAF inhibitor and MEK1/2 inhibitor secondary assays, (c) Venn diagram representing the overlap in ORFs scoring in Torinl and BEZ-235 secondary assays.

[0020] Supplementary Figure 10: GI50 values for BRAFV600E mutant melanoma cell lines treated with MAPK inhibitors targeting RAF, MEK1/2, or ERK (Ref. 15).

[0021] Supplementary Figure 11: Overexpression of IKBKB in A375 cells confers selective resistance to MAPK pathway inhibitors but not to mTOR pathway inhibitors relative to a neutral control (MEKl). Measurements were taken at a single concentration of each drug (GDC -0879 = ΙμΜ, AZD-6244 = 250 nM, Torin-1 = 200 nM, and BEZ-235 = 200 nM).

[0022] Supplementary Figure 12: IKBKB overexpression or exogenous TNFi (25 ng/mL) confer resistance to PLX4720 that is comparable in magnitude to C-RAF overexpression. MEKl overexpression serves as a neutral control, (a) Fold-change in PLX4720 GI50 following TNFa addition, IKBKB overexpression, or C-RAF overexpression in A375 and UACC-62 cell lines, (b) Overexpression of wild-type MEKl has no effect on PLX4750 GI50 relative to parental cell lines.

[0023] Supplementary Figure 13: Resistance mediated by overexpression of IKBKB or TRAF2 or addition of exogenous TNFa is associated with phosphorylation of RelA, a commonly used indicator of NF-κΒ activity, and is at least partially independent of ERK reactivation, (a) A375 cells, (b) Sk-Mel-28 cells.

[0024] Supplementary Figure 14: Resistance is associated with a reversal of PLX4720- induced apoptosis but not a bypass of cell cycle arrest. Apoptosis, indicated by Annexin V (+) / propidium iodide (PI) (-) staining, is induced by PLX4720 treatment and partially reversed by the addition of 25 ng/mL TNFa (left). Cell cycle phase distributions are unaltered (right), (a) A375 cells, (b) Colo679 cells.

Detailed Description of Certain Embodiments of the Invention 10025| I. Definitions

[00261 "Adherent cells" are cells that are capable of adhering to a suitable surface and, in general, have a tendency to adhere rather than remaining floating in suspension. Typically, adherent cells survive and proliferate more successfully while attached to a surface rather than floating in suspension (under conditions that are otherwise comparable and otherwise suitable for survival and proliferation). As known in the art, adherent cells include most cell types obtained from solid organs and cell lines derived from such cells that have not been adapted or selected for ability to survive and proliferate in suspension culture. For example, most fibroblast and epithelial cells and cell lines are adherent cells. In some cases, adherent cells require a surface suitable for cell adhesion in order to survive and/or proliferate to any significant extent.

[0027] The term "agent" encompasses any entity the effect of which on cell phenotype is of interest and that can be deposited on, synthesized on, or otherwise physically associated with a surface. An agent can comprise or consist of, e.g., a nucleic acid, a polypeptide, a virus, or a small molecule. In many embodiments, an agent comprises a vector that transfers a nucleic acid into a cell, wherein the effect of the nucleic acid or a product encoded by the nucleic acid on cell phenotype is of interest. In some embodiments the nucleic acid comprises a template for transcription of one or more RNAs by the cell, wherein the effect of the RNA(s) or protein(s) encoded thereby on cell phenotype is of interest. In some embodiments, the nucleic acid (or a DNA reverse transcribed therefrom within the cell) at least in part integrates into the host cell genome.

[0028] The term "array" refers to a multi-dimensional arrangement, typically a two dimensional arrangement, of features. Typically, an array of the present invention is an "addressable array, which term refers to an array wherein the identity of an agent or combination of agents, and/or concentration of agent(s) in most (e.g., at least 75%, 80%, 85%, 90%, 95%, or more) or all of the features can be determined by the spatial location of the feature in the array, or wherein the spatial location of a feature can be used to determine the identity of the agent(s) associated with that feature, or wherein given the identity of an agent the features containing that agent are known based on their spatial location. Typically, an array of the present invention is an addressable array. Typically features of an array are separated from adjacent features by regions that are substantially devoid of agents whose effect on cell phenotype is to be assessed.

[0029| The term "biodegradable" with respect to a material means that the material loses stability or physical integrity over time when subjected to a biological environment, such as under cell culture conditions. For example, a matrix composed of a biodegradable material may physically erode or chemically degrade and become smaller or less dense over time and, potentially, eventually disappear.

|0030] The term "biocompatible" as used herein with respect to a material means that neither the material nor its degradation products (if any), are toxic or elicit an adverse biologic response in cultured cells or tissues in the context in which the material is used.

[0031| The term "cell adhesion", also referred to as "cell attachment" refers to a reasonably stable physical association of cells with a surface. Typically the association is at least sufficiently stable to withstand mild agitation or gentle removal (such as by pouring) of culture medium from a tissue culture dish.

[0032] A "cell adhesion resistant material" is a material that is not hospitable to adhesion of many diverse adherent mammalian cell types. For example, less than 20%, e.g., less than 10%, 5%, 2%, or 1% of cells that settle on a surface made of or coated with the material may adhere thereto. Untreated polystyrene is an example of a cell adhesion resistant material. Tissue culture dishes composed of polystyrene are typically subjected to any of various treatments such as corona discharge, γ-irradiation, or chemical treatment, to produce a surface that is suitable for culturing a wide variety of adherent eukaryotic cells.

[0033] The term "cell adhesive material" refers to a material that promotes adhesion of many diverse mammalian cell types. The presence of a cell adhesive material on a region of a surface may, for example, result in a greater density of cells on the region that comprises the cell adhesive material than on regions that lack the cell adhesive material or may result in a greater percentage of cells remaining adherent to the region for a time period of interest than remain adherent to regions that lack the cell adhesive material over that time period.

[0034] The term "ceil migration" refers to the movement (translocation) of cells from one location of a surface to another. It will be understood that cell migration (movement of cells over a surface) can entail formation of new contacts with the surface (e.g., focal adhesions) at the "front" or "leading edge" of the cell and the release of contacts at the "trailing" edge of the cell, while the cell as a whole remains attached to the surface.

[0035] "Conditions suitable for cell adhesion" refer to suitable culture conditions for cells of that type, including use of suitable culture medium, a suitable temperature, humidity, and other conditions suitable for maintaining cell viability and, often, suitable for cell proliferation.

[0036] "Detectable label" is used herein to refer to an entity that can be detected, e.g., using optical, electrical, chemical, spectroscopic, biochemical, immunochemical, photochemical, and/or magnetic means. Suitable detectable labels of use in various embodiments of the invention include, but are not limited to, luminescent agents (e.g., bioluminescent or chemiluminescent proteins), fluorescent agents (e.g., fluorescent proteins), enzymes, and affinity tags (e.g., epitope tags). Often, a detectable label is an entity that generates a signal that can be measured and whose intensity is related to the amount of label (e.g., number of molecules) present (e.g., in a sample). A detectable label may be directly detectable (i.e., it can be detected without requiring binding to or reaction with other molecule(s) and/or it may be indirectly detectable (i.e., it is made detectable through interaction with (e.g., reaction or binding to) another entity that is detectable (either directly or indirectly). For example, a fluorescent or radioactive substance is generally directly detectable. Many labels are detectable following interaction with a substrate, e.g., luciferases are detectable based on their catalysis of a reaction that produces light. An epitope tag is often detected following binding of an antibody comprising a directly detectable moiety such as a fiuorophore or following binding of an antibody conjugated to an enzyme, wherein the enzyme reacts with a substrate to generate a signal. Epitope tags are often short peptide sequences to which high-affinity antibodies exist (or can be readily produced). In some embodiments, an epitope tag consists of a sequence that is not present in endogenous proteins expressed by a particular cell type or species (e.g., human cells). For example the genome of the cell may lack sequences encoding the tag or, if present, such sequences are not expressed. Examples of affinity tags include the FLAG, HA, TAP, Myc, and V5 tags, and 6XHis tags. Sequences of these tags, and suitable antibodies or other reagents for detecting them and/or for isolating proteins labeled with a particular tag, are well known in the art. Any of a wide variety of fluorescent or luminescent proteins may be used as detectable labels in various embodiments. Such proteins are well known in the art. Fluorescent proteins include, e.g., green fluorescent protein (GFP) from from the jellyfish Aequorea victoria, related naturally occurring green fluorescent proteins, and related proteins comprising chromophores that emit light of different colors such as red, yellow, and cyan. Many of these proteins are found in diverse marine animals such as Hydrozoa and Anthozoa species, crustaceans, comb jellies, and lancelets. See, e.g., See, e.g., Chalfie, M. and Kain, SR. (eds.) Green fluorescent protein: properties, applications, and protocols (Methods of biochemical analysis, v. 47). Wiley - Interscience, Hoboken, N.J., 2006, which discusses GFP and numerous other fluorescent or luminescent proteins. See also Chudakov, DM, et al., Physiol Rev. 90(3): 1103-63, 2010, for further information and references. In some embodiments, a detectable label comprises a monomeric fluorescent protein. Non-limiting examples of monomeric fluorescent proteins include Sirius, Azurite, EBFP2, TagBFP, mTurquoise, ECFP, Cerulean, TagCFP, mTFPl, mUkGl, mAGl, AcGFPl, TagGFP2, EGFP, mWasabi, EmGFP, TagYPF, EYFP, Topaz, SYFP2, Venus, Citrine, mKO, mK02, mOrange, mOrange2, TagRFP, TagRFP-T, mStrawberry, mRuby, mCherry, mRaspberry, mKate2, mPlum, mNeptune, T- Sapphire, mAmetrine, mKeima. See Chudakov DM (cited above). In some embodiments, a detectable label comprises a luciferase. As known in the art, "luciferase" refers to members of a class of enzymes that catalyze reactions that result in production of light. Luciferases are found in a variety of organisms including a variety of marine copepods, beetles, and others, and a number of these proteins have been cloned. Examples of luciferases include, e.g., luciferase from species of the genus Renilla (e.g., Renilla reniformis (Rluc), or Renilla mulleri luciferase), luciferase from species of the genus Gaussia or Metridia, luciferase from species of the genus Pleuromamma, beetle luciferases (e.g. luciferase of the firefly Photinus pyralis or of the Brazilian click beetle), etc. In certain embodiments of the invention in which a secreted protein is used as a detectable label and the naturally occurring form contains a signal sequence effective to direct secretion of the luciferase when expressed in eukaryotic cells (e.g. mammalian cells), the signal sequence may be at least in part removed or modified so that is no longer functional in the cells to be used in the assay. In some embodiments, a fluorescent or luminescent protein or luciferase is an engineered variant of a naturally occurring protein. Such variants may, for example, have increased stability (e.g., increased photostability, increased pH stability), increased fluorescence or light output, reduced tendency to dimerize, oligomerize, or aggregate, an altered absorption/emission spectrum (in the case of a fluorescent protein) and/or an altered substrate utilization. See, e.g., Chalfie, M. and Kain, SR (cited above) for examples. For example, the A. Victoria GFP variant known as enhanced GFP (eGFP) may be used. A variant of a naturally occurring luciferase that provides higher light output than the naturally occurring form and/or utilizes a coelentarazine analog as a substrate can be used. See, e.g., Loening, AM, et al., Protein Engineering, Design and Selection (2006) 19 (9): 391-400, for examples with respect to Renilla luciferase. In some embodiments, a nucleic acid sequence encoding a detectable protein (e.g., GFP, luciferase, etc.) is codon-optimized for expression in eukaryotic cells, e.g., cells that may be seeded onto an array. For example, the sequence may be codon-optimized for expression in mammalian cells, e.g., human cells.

[0037] The term "feature" as used herein in describing an array of the invention, refers to a discrete area of a surface that comprises a homogenous collection of an agent (or agents in certain embodiments). It will be understood that "homogeneous" in this context refers to the identity of the agent and does not necessarily imply that the agent is uniformly distributed over the area of a feature, although in certain embodiments this may be the case. The term "feature" is intended to include any geometry that permits placement of agents at discrete defined locations. A feature may have substantial length, width, and, in some embodiments, depth. Exemplary features are often relatively flat, and may have, for example, circular, oval, square, rectangular, or triangular shapes. In some embodiments, features project above a substantially planar surface. In some embodiments, features are at least partly located within depressions or cavities that project below a substantially planar surface. In some embodiments, features have substantial depth in the vertical dimension. For example, features can be rounded, domed, or approximately cylindrical, spherical, cuboidal, or right- angled parallelepiped shaped. Features can be considered to be different (distinct) from one another or the same as one another based on the agent(s) contained therein. One feature can be different than another feature if the agents of the different features have different chemical identities (e.g., different sequences or chemical structures) or different concentrations or amounts. In the case of features that comprise multiple agents, one feature can be different from another feature if at least one agent is present in one feature and is either absent from the other feature or is present in a different amount or concentration. In some embodiments, features can differ from one another based on the identity or amount of one or more non- agent component(s) of said features. For example, features can differ from one another based on their comprising different cell adhesive materials.

[0038] The term "loss of function" as it refers to the effect of an agent on a cell, refers to those agents that, when present or expressed in a cell, inhibit expression of a gene or otherwise render the gene product thereof to have substantially reduced activity, or essentially no activity relative to one or more activities of the gene product under substantially identical conditions in the absence of the agent. In some embodiments substantially reduced activity refers to a reduction of at least 50%, e.g., a reduction of between 50% and 80%, or between 80% and 95%, or even greater reduction. In some embodiments "essentially no activity" refers to an activity level that is not significantly different from background level that reflects complete loss of gene function or performing an assay under conditions in which the gene product is absent. A complete loss of gene function may be achieved, for example, by deleting a gene. In some embodiments, a "partial loss of function" refers to inhibition by less than 50%, e.g., inhibition by 10%, 20%, 30%, 40%, up to 50%.

[0039] The term "modulator" refers to an entity (e.g., an agent, test compounds, or other entity) or condition that alters, e.g., inhibits (reduces, decreases, suppresses, or represses) or enhances (activates, stimulates, increases, or promotes), a phenotype, process, pathway, phenomenon, state, or activity. For example, a modulator of protein activity may increase or decrease the level of one or more activit(ies) of a protein, e.g., an enzymatic activity. The term "modulate" refers to altering, e.g., inhibiting or enhancing a phenotype, process, pathway, phenomenon, state, or activity. As another example, a modulator of cell proliferation may increase or decrease cell proliferation.

[0040] The terms "protein", "polypeptide", and "peptide" are used interchangeably herein. In some embodiments, a "peptide" is a relatively short polypeptide chain (e.g., up to about 60 amino acids long, e.g., between 3 and 25 amino acids long). In some embodiments, a "protein" comprises multiple polypeptide chains which may be covalently or noncovalently associated with each other.

[0041] The terms "nucleic acid" and "polynucleotide" are used interchangeably herein. In various embodiments the term "nucleic acid" encompasses naturally occurring polymers of nucleosides, such as DNA and RNA, and non-naturally occurring polymers of nucleosides or nucleoside analogs. In some embodiments a nucleic acid comprises standard nucleosides (abbreviated A, G, C, T, U). In other embodiments a nucleic acid comprises one or more non-standard nucleosides. In some embodiments, one or more nucleosides are non-naturally occurring nucleosides or nucleotide analogs. A nucleic acid can comprise modified bases (for example, methylated bases), modified sugars (2'-fluororibose, arabinose, or hexose), modified phosphate groups or other linkages between nucleosides or nucleoside analogs (for example, phosphorothioates or 5'-N-phosphoramidite linkages), locked nucleic acids, or morpholinos. In some embodiments, a nucleic acid comprises nucleosides that are linked by phosphodiester bonds, as in DNA and R A. In some embodiments, at least some nucleosides are linked by non-phosphodiester bond(s). A nucleic acid can be single-stranded, double-stranded, or partially double-stranded. An at least partially double- stranded nucleic acid can have one or more overhangs, e.g., 5' and/or 3' overhang(s). Nucleic acid modifications (e.g., nucleoside and/or backbone modifications, including use of non-standard nucleosides) known in the art as being useful in the context of RNA interference (RNAi), aptamer, or anti sense-based molecules for research or therapeutic purposes are contemplated for use in various embodiments of the instant invention. See, e.g., Crooke, ST (ed.) Antisense drug technology: principles, strategies, and applications, Boca Raton: CRC Press, 2008; Kurreck, J. (ed.) Therapeutic oligonucleotides, RSC biomolecular sciences.

Cambridge: Royal Society of Chemistry, 2008. In some embodiments, a modification increases half-life and/or stability of a nucleic acid, e.g., in a tissue culture environment, relative to RNA or DNA of the same length and strandedness. In some embodiments, between 5% and 95% of the nucleosides in one or both strands of a nucleic acid is modified. Modifications may be located uniformly or nonuniformly, and the location of the modifications (e.g., near the middle, near or at the ends, alternating, etc.) can be selected to enhance desired propert(ies). A nucleic acid may comprise a detectable label, e.g., a fluorescent dye, radioactive atom, etc. "Oligonucleotide" refers to a relatively short nucleic acid, e.g., typically between about 4 and about 60 nucleotides long.

100421 "Profile" refers to a collection of information regarding an entity or entit(ies). For example, a profile may represent the extent to which an entity or entit(ies) exhibit(s) various characteristics or activities of interest, A profile of a compound ("compound profile") may include information regarding the extent to which the compound binds to or affects the activity of each of a plurality of proteins (e.g., as determined using a particular assay). A profile of a protein may include information regarding the extent to which it is bound by each of a plurality of compounds and/or the extent to which an activity of the protein is modulated (e.g., inhibited or activated) by each of a plurality of compounds. In some embodiments, a profile includes quantitative information (e.g., measurements of a characteristic or activity of interest). A profile may, if desired, be presented or displayed in any of variety of formats, e.g., lists, tables, graphs, charts, plots, heatmaps, dendrograms, etc. "Profiling" refers to the process of acquiring the information (e.g., by performing one or more assays) and, optionally, processing and/or analyzing the information acquired.

[0043] "Purified" refers to entities that have been separated from most of the components with which they are associated in nature or when originally generated. In general, such purification involves action of the hand of man. Purified entities may be partially purified, substantially purified, or pure. Such entities may be, for example, at least 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or more than 99% pure. In some embodiments, a nucleic acid or polypeptide is purified such that it constitutes at least 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or more, of the total nucleic acid or polypeptide material, respectively, present in a preparation. Purity can be based on, e.g., dry weight, size of peaks on a chromatography tracing, molecular abundance, intensity of bands on a gel, or intensity of any signal that correlates with molecular abundance, or any art- accepted quantification method. In some embodiments, water, buffers, ions, and/or small molecules (e.g., precursors such as nucleotides or amino acids), can optionally be present in a purified preparation. A purified molecule may be prepared by separating it from other substances (e.g., other cellular materials), or by producing it in such a manner to achieve a desired degree of purity. In some embodiments, a purified molecule or composition refers to a molecule or composition that is prepared using any art-accepted method of purification. In some embodiments "partially purified" means that a molecule produced by a cell is no longer present within the cell, e.g., the cell has been lysed and the molecule has been separated or segregated from at least some molecules of the same type (protein, RNA, DNA, etc.) that were present in the lysate.

[0044] The term "recombinant protein" refers to a polypeptide which is produced by recombinant DNA techniques, wherein generally, DNA encoding the polypeptide is inserted into a suitable expression vector which is introduced into a cell (sometimes termed a "host cell") to produce the protein. It will be understood that the nucleic acid may be, and often is, copied by the cell and inherited by descendants of the cell into which the nucleic acid was introduced, which cells can also produce the recombinant protein.

[0045] The term "replicates", when used herein to refer to features, refers to features that are essentially identical. Features are "essentially identical" if they contain the same agent(s) and approximately the same concentration or amount thereof and the same non-agent component(s) (if any). "Replicates" when used herein to refer to arrays, refers to arrays having the same features located thereon, in the same configuration. In some embodiments, replicate arrays will have been manufactured in a single printing "run" or from a single preparation of a set of agents. In some embodiments, replicate arrays will have been manufactured in multiple printing "runs" or from different preparations of a set of agents.

[0046] "RNA interference" (RNAi) refers to a phenomenon whereby double-stranded RNA (dsRNA) triggers the sequence-specific degradation or translational repression of a corresponding mRNA having complementarity to a strand of the dsRNA. It will be appreciated that the complementarity between the strand of the dsRNA and the mRNA need not be 100% but need only be sufficient to mediate inhibition of gene expression (also referred to as "silencing" or "knockdown"). For example, the degree of complementarity can be such that the strand can either (i) guide cleavage of the mRNA in the RNA-induced silencing complex (RISC); or (ii) cause translational repression of the mRNA. In certain embodiments the double-stranded portion of the RNA is less than about 30 nucleotides in length, e.g., between 17 and 29 nucleotides in length. In certain embodiments a first strand of the dsRNA is at least 80%, 85%, 90%, 95%, or 100% complementary to a target mRNA and the other strand of the dsRNA is at least 80%, 85%, 90%, 95%, or 100% complementary to the first strand. In mammalian cells, RNAi may be achieved by introducing an appropriate double-stranded nucleic acid into the cells or expressing a nucleic acid in cells that is then processed intracellularly to yield dsRNA therein. Nucleic acids capable of mediating RNAi are referred to herein as "RNAi agents". Exemplary nucleic acids capable of mediating RNAi are a short hairpin RNA (shRNA), a short interfering RNA (siRNA), and a microRNA precursor. These terms are well known in the art. siRNAs typically comprise two separate nucleic acid strands that are hybridized to each other to form a duplex. They can be synthesized in vitro, e.g., using standard nucleic acid synthesis techniques. siRNAs are typically double-stranded oligonucleotides having 16-30, e.g., 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30 nucleotides (nt) in each strand, wherein the double-stranded oligonucleotide comprises a double-stranded portion between 15 and 29 nucleotides long and either or both of the strands may comprise a 3' overhang between, e.g., 1-5 nucleotides long, or either or both ends can be blunt. In some embodiments, an siRNA comprises strands between 19 and 25 nt, e.g., between 21 and 23 nucleotides long, wherein one or both strands comprises a 3' overhang of 1-2 nucletides. One strand of the double-stranded portion of the siRNA (termed the "guide strand" or "antisense strand") is substantially complementary (e.g., at least 80% or more, e.g., 85%, 90%, 95%, or 100%) complementary to (e.g., having 3, 2, 1, or 0 mismatched nucleotide(s)) a target region in the mRNA, and the other double-stranded portion is substantially complementary to the first double-stranded portion. In many embodiments, the guide strand is 100% complementary to a target region in an mRNA and the other passenger strand is 100% complementary to the first double-stranded portion (it is understood that, in various embodiments, the 3' overhang portion of the guide strand, if present, may or may not be complementary to the mRNA when the guide strand is hybridized to the mRNA). In some embodiments, a shRNA molecule is a nucleic acid molecule comprising a stem-loop, wherein the double-stranded stem is 16-30 nucleotides long and the loop is about 1-10 nucleotides long. siRNA can comprise a wide variety of modified nucleosides, nucleoside analogs and can comprise chemically or biologically modified bases, modified backbones, etc. Without limitation, any modification recognized in the art as being useful for RNAi can be used. Some modifications result in increased stability, cell uptake, potency, etc. Some modifications result in decreased immunogenicity or clearance. In certain embodiments the siRNA comprises a duplex about 19-23 (e.g., 19, 20, 21, 22, or 23) nucleotides in length and, optionally, one or two 3' overhangs of 1 -5 nucleotides in length, which may be composed of deoxyribonucleotides. shRNA comprise a single nucleic acid strand that contains two complementary portions separated by a predominantly non- selfcomplementary region. The complementary portions hybridize to form a duplex structure and the non-selfcomplementary region forms a loop connecting the 3' end of one strand of the duplex and the 5' end of the other strand. shRNAs undergo intracellular processing to generate siRNAs. Typically, the loop is between 1 and 8, e.g., 2-6 nucleotides long.

[0047] MicroRNAs (miRNAs) are small, naturally occurring, non-coding, single- stranded RNAs of about 21-25 nucleotides (in mammalian systems) that inhibit gene expression in a sequence-specific manner. They are generated intracellularly from precursors (pre-miRNA) having a characteristic secondary structure comprised of a short hairpin (about 70 nucleotides in length) containing a duplex that often includes one or more regions of imperfect complementarity which is in turn generated from a larger precursor (pri-miRNA). Naturally occurring miRNAs are typically only partially complementary to their target mRNA and often act via translational repression. RNAi agents modelled on endogenous miRNA or miRNA precursors are of use in certain embodiments of the invention. For example, an siRNA can be designed so that one strand hybridizes to a target mRNA with one or more mismatches or bulges mimicking the duplex formed by a miK A and its target mRNA. Such siRNA may be referred to as artificial miR A or miRNA-like molecules. Artificial miRNA may be encoded by precursor nucleic acids whose structure mimics that of naturally occurring miRNA precursors.

[0048] In certain embodiments an RNAi agent is a vector (e.g., a plasmid or virus) that comprises a template for transcription of an siRNA (e.g., as two separate strands that can hybridize to each other), shRNA, or microRNA precursor. Typically the template encoding the siRNA, shRNA, or miRNA precursor is operably linked to expression control sequences (e.g., a promoter), as known in the art. Such vectors can be used to introduce the template into vertebrate cells, e.g., mammalian cells, and result in transient or stable expression of the siRNA, shRNA, or miRNA precursor. Precurors (shRNA or miRNA precursors) are processed intracellularly to generate siRNA or miRNA.

[0049] In general, small RNAi agents such as siRNA can be chemically synthesized or can be transcribed in vitro or in vivo from a DNA template either as two separate strands that then hybridize, or as an shRNA which is then processed to generate an siRNA. Often RNAi agents, especially those comprising modifications, are chemically synthesized.

[0050] The terms "sequence of interest" and "nucleic acid of interest" refer to the portion or portions of a nucleic acid introduced into the eukaryotic cells (e.g., via a vector such as a virus or plasmid) that is of interest with respect to its ability to confer a change in the phenotype of the cells. In general, though not always, if the agents comprise nucleic acids, the sequence of interest will be that portion(s) of the nucleic acids that is varied from one feature of an array to the next. Thus in many embodiments of the invention the sequence of interest will be that portion of the nucleic acid contained in a virus vector that is varied from one feature of the array to the next. The sequence of interest can be, for example, a coding sequence for a protein, a "coding" sequence for an RNA molecule (e.g., which is transcribed into an anti-sense RNA sequence, a ribozyme or an at least partly double-stranded RNA such as a shRNA), or a regulatory sequence (e.g., as part of a reporter gene construct).

[0051] The term "selectable marker" is used herein to refer to a biomolecule, typically a protein, that confers a growth advantage or disadvantage on cells that express it as compared with cells that do not express it. By "growth advantage" is meant either enhanced viability (e.g., cells with a growth advantage have an increased average life span, relative to otherwise identical cells), increased rate of cell proliferation relative to otherwise identical cells, or both. In other words, cells that express the selectable marker have an increased or decreased ability to survive or proliferate as compared with cells that do not express the selectable marker. Often, the growth advantage or disadvantage is significant or evident only under particular conditions ("selective conditions"). The selective conditions are often the presence or absence of a particular compound in the culture medium, although environmental conditions such as increased temperature, increased radiation, etc., may serve as selective conditions. In many embodiments, a selectable marker is a "positive selectable marker", meaning that cells expressing the marker have a selective advantage (e.g., under selective conditions). In some embodiments a selectable marker is a "negative selection marker", which confers a disadvantage on cells that express it (e.g., under selective conditions). One of ordinary skill in the art will be aware of numerous selectable markers. Examples of positive selection markers include, e.g., enzymes that confer resistance to various compounds (e.g., antibiotics) that are otherwise deleterious to cells and biosynthetic enzymes that allow cells to survive or proliferate in the absence of particular nutrients in the culture medium. In the presence of an appropriate concentration of antibiotic (selective conditions), such a marker confers a growth advantage on a cell that expresses the marker. Thus cells that express the antibiotic resistance marker are able to survive and/or proliferate in the presence of the antibiotic (e.g., by inactivating the antibiotic) while cells that do not express the antibiotic resistance marker are not able to survive and/or are unable to proliferate in the presence of the antibiotic. For example, a selectable marker of this type commonly used in mammalian cells is the neomycin resistance gene (an aminoglycoside 3'-phosphotransferase). Expression of this selectable marker renders cells resistant to various antibiotics such as G418. Additional selectable markers of this type include enzymes conferring resistance to zeocin™, hygromycin, blasticidin, puromycin, etc. For example, puromycin resistance can be conferred by a puromycin N-acetyl-transferase. These enzymes and the genes encoding them are well known in the art and are available in numerous vectors. Another class of selectable markers is nutritional markers. Such markers are generally enzymes that function in a biosynthetic pathway to produce a compound that is needed for cell growth or survival. In general, under nonselective conditions the required compound is present in the environment or is produced by an alternative pathway in the cell. Under selective conditions, functioning of the biosynthetic pathway in which the marker is involved is needed to produce the compound.

[0052] The term "small molecule" refers to an organic molecule that is less than about 2 kilodaltons (KDa) in mass. In some embodiments, the small molecule is less than about 1.5 KDa, or less than about 1 KDa. In some embodiments, the small molecule is less than about 800 daltons (Da), 600 Da, 500 Da, 400 Da, 300 Da, 200 Da, or 100 Da. Often, a small molecule has a mass of at least 50 Da. In some embodiments, a small molecule contains multiple carbon-carbon bonds and can comprise one or more heteroatoms and/ or one or more functional groups important for structural interaction with proteins (e.g., hydrogen bonding), e.g., an amine, carbonyl, hydroxyl, or carboxyl group, and in some embodiments at least two functional groups. Small molecules often comprise one or more cyclic carbon or heterocyclic structures and/or aromatic or polyaromatic structures, optionally substituted with one or more of the above functional groups. In some embodiments a small molecule is an artificial (non-naturally occurring) molecule. In some embodiments, a small molecule is non- polymeric. In some embodiments, a small molecule is not an amino acid. In some embodiments, a small molecule is not a nucleotide. In some embodiments, a small molecule is not a saccharide. In some embodiments, the term "small molecule" specifically excludes molecules that are components of a tissue culture medium.

[0053] The term "test compound" may be used interchangeably herein with "compound of interest" (sometimes simply referred to as a "compound") and can refer to any molecular or supramolecular entity (e.g., a small molecule, polypeptide, nucleic acid, lipid, polysaccharide, virus, cell, etc.) or combination thereof. In many embodiments, a test compound is an entity whose effect on cell phenotype is to be assessed (or is being or has been assessed) using an array of the invention. A test compound may be a therapeutic agent or candidate therapeutic agent or a hit or lead compound under consideration for further drug development efforts. The identity (e.g., chemical formula, name, sequence, etc.) of a test compound may or may not be known in various embodiments. It will be understood that a test compound may be present in addition to any substances normally found in cell culture medium, and in addition to a compound that is used for selection purposes, such as an antibiotic.

[0054| The term "tumor" is used herein generally interchangeably with "cancer". In many embodiments a tumor is a malignant growth, which may be metastatic or non- metastatic in various embodiments. Cancer, as used herein, encompasses malignant solid tumors (carcinomas, sarcomas) and hematologic malignancies. As used herein, the term cancer includes, but is not limited to, the following types of cancer: breast cancer; biliary tract cancer; bladder cancer; brain cancer including glioblastomas and medulloblastomas; cervical cancer; choriocarcinoma; colon cancer; endometrial cancer; esophageal cancer; gastric cancer; hematological neoplasms including acute lymphocytic leukemina and acute myelogenous leukemia; T-cell acute lymphoblastic leukemia lymphoma; hairy cell leukemia; chronic lymphocytic leukemia, chronic myelogenous leukemia, multiple myeloma; AIDS- associated leukemias and adult T-cell leukemia/lymphoma; intraepithelial neoplasms including Bowen's disease and Paget's disease; liver cancer; lung cancer; lymphomas including Hodgkin's disease and lymphocytic lymphomas; neuroblastomas; oral cancer including squamous cell carcinoma; ovarian cancer including those arising from epithelial cells, stromal cells, germ cells and mesenchymal cells; pancreatic cancer; prostate cancer; rectal cancer; sarcomas including leiomyosarcoma, rhabdomyosarcoma, liposarcoma, fibrosarcoma, and osteosarcoma; skin cancer including melanoma, Merkel cell carcinoma, Kaposi's sarcoma, basal cell carcinoma, and squamous cell cancer; testicular cancer including germinal tumors such as seminoma, non-seminoma (teratomas, choriocarcinomas), stromal tumors, and germ cell tumors; thyroid cancer including thyroid adenocarcinoma and medullar carcinoma; and renal cancer including adenocarcinoma and Wilms tumor. In some embodiments the term "tumor" encompasses proliferative diseases such as

myeloproliferative disease, e.g., myelodysplastic syndrome, myelofibrosis, essential thrombocythemia, or polycythemia vera. In some embodiments, a proliferative disease is neurofibromatosis, tuberous sclerosis, or lymphangioleiomyomatosis. Information regarding cancer and treatment of cancer may be found, e.g., in Cancer: Principles and Practice of Oncology (V.T. DeVita et al., eds., J.B. Lippincott Company, 8th ed., 2008 or 9th ed., 2011.

[0055] A "variant" of a particular polypeptide refers to a polypeptide that differs from such polypeptide (sometimes referred to as the "original polypeptide") by one or more amino acid alterations, e.g., addition(s), deletion(s), and/or substitution(s). Sometimes an original polypeptide is a naturally occurring polypeptide (e.g., from human or non-human animal) or a polypeptide identical thereto. Variants may be naturally occurring or created using, e g., recombinant DNA techniques or chemical synthesis. An addition can be an insertion within the polypeptide or an addition at the N- or C-terminus. In some embodiments, the number of amino acids substituted, deleted, or added can be for example, about 1 to 30, e.g., about 1 to 20, e.g., about 1 to 10, e.g., about 1 to 5, e.g., 1 , 2, 3, 4, or 5. In some embodiments, a variant comprises a polypeptide whose sequence is homologous to the sequence of the original polypeptide over at least 50 amino acids, at least 100 amino acids, at least 150 amino acids, or more, up to the full length of the original polypeptide (but is not identical in sequence to the original polypeptide), e.g., the sequence of the variant polypeptide is at least 50%, 60%, 70%, 75%, 80%, 85%, 90%, 95%, 96%, 97%, 98%, 99%, or more identical to the sequence of the original polypeptide over at least 50 amino acids, at least 100 amino acids, at least 150 amino acids, or more, up to the full length of the original polypeptide. In some embodiments, a variant comprises a polypeptide at least 50%, 60%, 70%, 75%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, 99.5% or more identical to an original polypeptide over at least 50%, 60%, 70%, 80%, 85%, 90%, 91%, 92%, 93%, 94%, 95%, 96%, 97%, 98%, 99%, or 100% of the length of the original polypeptide. A variant may be a fusion protein (e.g., a recombinant fusion protein, a fusion protein arising as a result of a chromosomal translocation), a naturally occurring or engineered mutant protein, etc.

[0056] The term "vector" is used herein to refer to a nucleic acid or a virus or portion thereof (e.g., a viral capsid or genome) capable of mediating entry of, e.g., transferring, transporting, etc., a nucleic acid molecule into a cell. Where the vector is a nucleic acid, the nucleic acid molecule to be transferred is generally linked to, e.g., inserted into, the vector nucleic acid molecule. A nucleic acid vector may include sequences that direct autonomous replication (e.g., an origin of replication), or may include sequences sufficient to allow integration of part or all of the nucleic acid into host cell DNA. Useful nucleic acid vectors include, for example, DNA or RNA plasmids, cosmids, and naturally occurring or modified viral genomes or portions thereof or nucleic acids (DNA or RNA) that can be packaged into viral capsids, Plasmid vectors typically mclude an origin of replication and one or more selectable markers. Plasmids may include part or all of a viral genome (e.g., a viral promoter, enhancer, processing or packaging signals, etc.). Viruses or portions thereof that can be used to introduce nucleic acid molecules into cells may be referred to as viral vectors. Useful viral vectors include adenoviruses, adeno-associated viruses, retroviruses, lentiviruses, vaccinia virus and other poxviruses, herpesviruses (e.g., herpes simplex virus), and others. Viral vectors may or may not contain sufficient viral genetic information for production of infectious virus when introduced into host cells, i.e., viral vectors may be replication- defective. Where sufficient information is lacking it may, but need not be, supplied by a host cell or by another vector introduced into the cell in some embodiments. The nucleic acid to be transferred may be incorporated into a naturally occurring or modified viral genome or a portion thereof or may be present within the vims or viral capsid as a separate nucleic acid molecule. It will be appreciated that certain plasmid vectors that include part or all of a viral genome, typically including viral genetic information sufficient to direct transcription of a nucleic acid that can be packaged into a viral capsid and/or sufficient to give rise to a nucleic acid that can be integrated into the host cell genome and/or to give rise to infectious virus, are sometimes referred to in the art as viral vectors. Vectors may contain one or more nucleic acids encoding a marker suitable for use in the identifying and/or selecting cells that have or have not been transformed or transfected with the vector. Markers include, for example, proteins that increase or decrease either resistance or sensitivity to antibiotics (e.g.. an antibiotic-resistance gene encoding a protein that confers resistance to an antibiotic such as puromycin, hygromycin or blasticidin) or other compounds, enzymes whose activities are detectable by assays known in the art (e.g., β-galactosidase or alkaline phosphatase), and proteins or RNAs that detectably affect the phenotype of transformed or transfected cells (e.g., fluorescent proteins). Expression vectors are vectors that include regulatory sequence(s), e.g., expression control sequences such as a promoter, sufficient to direct transcription of an operably linked nucleic acid. Regulatory sequences may also include enhancer sequences or upstream activator sequences. Vectors may optionally include 5' leader or signal sequences. Vectors may optionally include cleavage and/or polyadenylation signals and/or a 3' untranslated region. Vectors often include one or more appropriately positioned sites for restriction enzymes, to facilitate introduction into the vector of the nucleic acid to be expressed. An expression vector comprises sufficient cis-acting elements for expression; other elements required or helpful for expression can be supplied by the host cell or in vitro expression system.

[0057] II, Arrays and Methods of Making Thereof

[0058] The present invention arose at least in part from the Applicants' recognition that there is a significant unmet need for technologies that facilitate rapid and cost-effective functional genomic screening across large numbers of mammalian cell lines or cell samples, culture conditions, cell perturbations (e.g., genetic or chemical perturbations), and/or assay formats. In some aspects, the invention provides arrays suitable for use in cell-based screens, wherein the arrays possess one or more of the following attributes; (1) compatibility with robust, quantitative, high-throughput screening; (2) compatibility with rapid, multiplexed detection of image- or proliferation-based readouts following either negative or positive selection; (3) compatibility with existing lentiviral open reading frame (ORF) and short hairpin RNA (shRNA) libraries; (4) compatibility with a broad range of adherent mammalian cell lines varying in regard to infectability, morphology, and migratory properties; (5) compatibility with semi-automated, scalable production and long-term storage; (6) suitable for use with small quantities of library constructs, cells, and detection reagents; (7) not requiring specialized screening facilities or fluid handling equipment; and (8) low cost relative to existing technologies based on microwell plates. Without limiting the invention in any way, the arrays may find particular use in screens in which it is desired to assess the effect of thousands, tens of thousands, or more different genetic and/or chemical perturbations on cell phenotype.

[0059] In some aspects, an array of the invention comprises multiple features arranged on a surface at distinct locations. The features can be produced using any of a variety of approaches, as described further below. For example, features can be produced by placing an agent, optionally in a composition containing one or more additional materials, onto a surface. For purposes of the invention, the act of placing an agent onto a surface may be referred to as "depositing" or "printing" the agent. Any of variety of different types of agents that are potentially capable of interacting with cells can be used to produce an array in various embodiments of the invention. For example, agents can comprise viruses, small molecules, nucleic acids, or proteins. The agent may be deposited in a liquid composition comprising one or more other substances. For purposes of the present invention such a composition may be referred to as a "printing buffer". An appropriate printing buffer may be selected based at least in part on the identity of the agent(s) to be deposited. Typically the buffer is an aqueous medium. The printing buffer may contain, for example, one or more buffering compounds, salts, stabilizers, osmotic agents, and/or substance(s) that promote viral infection or transfection. In many embodiments, multiple agents are deposited in a highly parallel manner such as by using a multipin printing device or multijet inket printer. In some embodiments, arrays containing mixtures of agents at each feature are constructed by, for example, mixing agents before printing, printing in serial, printing with masks, or printing with patterned printheads. For example, agents could be mixed in a container before printing and printed as a homogenous mixture. Alternatively, agents could be printed on top of one another or close to one another. Masks with different patterns of holes or print heads with different configurations could also be used to print multiple agents.

[0060] In general, in order to perform a screen, an array is seeded with adherent eukaryotic cells, some of which settle on features and adhere thereto. Cell seeding (also termed "plating" the cells) can be performed, for example, by adding cells to culture medium contained in a chamber in which the array is located. The cell-seeded array is maintained under conditions suitable for cell adhesion for an appropriate period of time to permit cells to adhere. An agent can interact with cells located on a feature containing the agent, resulting in an array of cell spots wherein at least some cells in each spot have interacted with the agent located at the feature to which they adhered. In many embodiments, an agent is reversibly affixed to the surface so that it can be released from the surface following cell seeding. Release of the agent can facilitate contact with overlying cells and potentially allow the agent to gain entry into a cell and, potentially, affect cell phenotype. After a suitable period of time, cells are assessed for one or more phenotype(s) of interest. For example, the effect of agent(s) on cell phenotype(s) can be determined and/or agent(s) that modulate a phenotype of interest can be identified. In some embodiments, a test compound is present in the culture medium during at least part of the maintenance period. In such embodiments, a cell phenotype of interest may be a response to the compound. In some embodiments agent(s) that modulate cell response to a compound may thus be identified. In general, detection of the effects of the agent (e.g., the effect of an introduced nucleic acid) on the cells can be performed using any of a variety of known techniques, some of which are discussed elsewhere herein.

[0061] In some aspects, the invention relates to the Applicants' recognition that mammalian cell lines can exhibit significant variability in regard to their tendency to migrate and/or in the size and localization of cell spots that form following seeding onto a surface. For example, as described in Example 1 , when arrays comprising lentiviruses deposited on standard, commercially available aminosilane-coated glass slides are seeded with mammalian cell lines, the size and localization of the region of infection corresponding to each feature can vary substantially from cell line to cell line, with some cell lines forming reasonably localized populations of infected cells around each feature and others forming diffuse, poorly localized regions of infected cells. The invention provides the insight that the propensity of mammalian cells to migrate, and the variability in size and localization exhibited by cell spots that form following cell seeding, can significantly limit both the achievable density of features and the range of cell lines compatible with cell microarray-based screening. In some aspects, the invention provides arrays that at least in part reduce these constraints. In some aspects, the invention provides microarrays that allow a wide range of cell-based screens to be performed without requiring optimization for different mammalian cell types. In some aspects, inventive arrays and screens provide results comparable to those obtained using microwell-based screening approaches, but screens can be performed using much lower medium volumes, smaller cell numbers, smaller amounts of test compounds, smaller amounts of detection reagents, and/or reduced cost per result obtained. [0062] In some aspects, the invention provides an array comprising a surface having multiple discrete features, wherein each feature comprises one or more agents to be contacted with eukaryotic cells, and wherein the array comprises means to confine eukaryotic cell adhesion to said features. In some embodiments, the array comprises means effective such that adherent mammalian cells are substantially confined to the features at a time point 24 hours after cell seeding. In some embodiments, mammalian cells are substantially confined to the features when assessed at a time point 36, 48, 60, 72, 96, 120, or 144, hours, or more, after seeding. In some embodiments, mammalian cells are substantially confined to the features when assessed at a time point 5-7 days, or 7-10 days, or more, after seeding. In some aspects, the invention provides an array comprising a surface having multiple distinct features, wherein each feature comprises an agent to be contacted with cells, and wherein the array comprises means effective to maintain the regions between features substantially devoid of cells at a time point 24 hours after seeding the array with adherent mammalian cells. In some embodiments, the regions between features are substantially devoid of cells when assessed at a time point 36, 48, 60, 72, 96, 120, or 144 hours, or more, after seeding. In some embodiments, the regions between features are substantially devoid of cells when assessed at a time point 5-7 days, or 7-10 days, or more, after seeding. It will be understood that typically, culture medium is gently removed from the array (or the array is removed from the culture medium) prior to assessment in order to help remove any cells that may have settled on the regions between the features but failed to adhere. In some embodiments, an array is gently agitated or rinsed one or more times (e.g., 2 or 3 times), e.g., with tissue culture medium or a physiologically acceptable solution such as phosphate buffered saline, prior to assessment.

[0063] In some embodiments, cells are considered to be "substantially confined to the features" if at least 80% of the cells located within a perimeter surrounding at least 80% of the features of an array are located on the features at the time of assessment and/or the density of cells located on the features is at least 4-fold higher than the density of cells in regions between the features at the time of assessment. In some embodiments, cells are considered to be "substantially confined to the features" if at least 85%, 90%, 95%, 96%, 97%, 98%, 99%, or more of the cells located within a perimeter surrounding at least 80% of the features of an array are located on the features at the time of assessment and/or the density of cells located on the features is at least 5, 6, 7, 8, 9, 10, 12, 15, 20, 25, 30, 40, or 50- fold higher than the density of cells in regions between the features at the time of assessment. In some embodiments, the regions between features are considered to be "substantially devoid of cells" if no more than 20% of the cells are located between the features at the time of assessment and/or the density of cells located between the features is at least 4-fold lower than the density of cells in regions between the features at the time of assessment. In some embodiments, regions between features are considered to be "substantially devoid of cells" if no more than 15%, 10%, 5%, 4%, 3%, 2%, 1%, or less of the cells located within a perimeter surrounding at least 80% of the features of an array are located between the features at the time of assessment and/or the density of cells located between the features is at least 5, 6, 7, 8, 9, 10, 12, 15, 20, 25, 30, 40, or 50-fold lower than the density of cells on the features at the time of assessment.

[0064] In some aspects, an array of the invention comprises a surface having multiple distinct features, wherein each feature comprises one or more agents to be contacted with cells, and wherein the array comprises means effective to inhibit mammalian cell migration away from said features. In many embodiments, the features are separated by regions that cause mammalian cells that adhere to a feature (and the descendants of such cells, if any) to remain substantially confined to that feature. In some embodiments, the features are separated by regions that inhibit mammalian cell migration. In many embodiments, regions that inhibit mammalian cell migration substantially surround the features up to their edges, so that mammalian cell migration away from the features and into the regions between features is substantially inhibited. In some embodiments, the regions that inhibit cell migration may not extend all the way up to the edges of the features. Instead, there may be a zone around the features that does not inhibit migration. The zone is surrounded by regions that inhibit cell migration, so that migration from one feature into an adjacent feature is substantially inhibited. In various embodiments, migration between features is considered substantially inhibited if no more than 2%, no more than 1%, no more than 0.5%, or no more than 0.1% of the cells adherent on a given feature (e.g., a control feature that does not comprise an agent), did not originate from cells deposited on that feature, e.g., when assessed at least 24, 48, 60, 72, 84, 96, 120 hours after plating cells onto an array.

[0065] In conventional microwell plates, cells are confined within distinct wells having walls that project upwards or downwards from a surface and enclose a volume, so that fluid and cells within each well is physically separated from fluid and cells in other wells and intermingling of fluid and cells cannot occur. In many embodiments of the instant invention, eukaryotic cells are confined to the locations of the features by means other than providing walls or barriers that project upward or downward from a surface and enclose a volume above each feature. In many embodiments, eukaryotic cells are confined to the locations of the features by means that do not require distinct physical structures such as wells or upwardly projecting walls to confine the cells.

[0066] In many embodiments, the regions that inhibit eukaryotic cell migration comprise a material that inhibits eukaryotic cell migration. The material that inhibits eukaryotic cell migration may possess physical, chemical, and/or biological properties that render it inhospitable to cell migration. In some embodiments, regions that inhibit eukaryotic cell migration comprise a cell adhesion resistant material. In some embodiments, regions that inhibit eukaryotic cell migration are substantially less amendable to mammalian cell adhesion or migration than an amiiiosilane-coated glass surface. In some embodiments an amino silane is gamma amino propyl silane (GAPS) or other another amino alkyl silane. In some embodiments, a first value is "substantially less" than a second value if the first value is less than 25% of the second value. In some embodiments, a first value is "substantially less" than a second value if the first value is less than 20%, 10%, 8%, 5%, 4%, 3%, 2%, or 1% of the second value.

[0067] In some embodiments, the features comprise a cell adhesive material. In some embodiments, features comprising a cell adhesive material are located on top of a surface that comprises a cell adhesion resistant material. The features thus form "islands" to which mammalian cells can adhere, while adhesion between features (and migration into the inter- feature regions) is inhibited. Mammalian cells thus remain substantially confined to the features, and regions between features remain substantially devoid of cells. In some embodiments, the features comprise a cell adhesive material having an agent located thereon. The agent may remain largely confined to the upper surface of the feature rather than being uniformly distributed within the cell adhesive material. In some embodiments, features that comprise a cell-adhesive material are substantially more hospitable to cell adhesion than an aminosilane-coated surface. In some embodiments, "substantially more" means more by a factor of at least 1.25-fold (i.e., 25% more), 1.5-fold (i.e., 50% more), or 2-fold, e.g., at least 3, 4, 5, 10, or 20-fold. In some embodiments, features that comprise a cell-adhesive material are at least 5, 10, or 20-fold more hospitable to cell adhesion than a cell adhesion resistant material located between the features. [0068] In some embodiments, an array of the invention comprises a substrate having a coating comprising a cell adhesion resistant material disposed on at least a portion of the surface of the substrate. The coating typically forms a layer that is relatively thin as compared with the length and width of the coated area. In many embodiments, the coating layer has a relatively uniform thickness. Coating can be performed using any of a variety of suitable methods. The particular method employed may be selected based at least in part on the particular surface and coating material. Exemplary methods include, e.g., spraying, dipping, vapor deposition (e.g., chemical or physical vapor deposition), spin coating, roller coating, in situ polymerization or self-assembly, lamination, etc. Slides coated with various materials are commercially available. In some embodiments, a commercially available coated substrate is used. In some embodiments an array of the invention is constructed on such a slide.

[0069] In some embodiments, multiple regions comprising a cell adhesive material are produced on top of the coating layer. Any of a variety of methods can be used to produce a surface having a pattern of cell-adhesive and cell adhesion resistant regions. For example, in various embodiments surface patterning can be achieved using contact or noncontact printing (e.g., using an appropriate pinhead), inkjet printing (e.g., using piezoelectric deposition), nanografting, self-assembly, dip-pen lithography, soft lithography, plasma polymerization, stamping, etc. In many embodiments, a standard micrarray printer is used. In some embodiments a printhead having solid pins is employed. In some embodiments, patterning is achieved using a mask that contains a pattern of holes or transparent areas. The mask may, e.g., cover and thereby protect regions of a surface that are not to be coated or to receive a particular treatment (e.g., a chemical treatment) while allowing other regions to be coated or to receive a treatment.

|0070] In some aspects, the invention provides methods of making an array. In some embodiments, an array is produced by a multistep process in which a substrate at least in part coated with a cell adhesion resistant material is provided, and then features are produced in two sequential steps: first, adhesive regions are formed by depositing a material that promotes localized ceil adhesion and, second, agents are deposited on top of the adhesive regions. The agent may, but need not, be disposed atop all or a substantial portion of the region of cell adhesive material on which it is deposited. For example, the agent may be concentrated around a central portion of a feature and may be surrounded by a zone of cell adhesive material that does not have an agent disposed thereon. In some embodiments, features are generated in a single step by depositing a composition that comprises a cell adhesive material and an agent on top of a surface that is at least in part cell adhesion resistant, e.g., a surface that at least in part comprises a cell adhesion resistant material (e.g., as a coating). Without limiting the invention, the first of the foregoing approaches may be advantageous in that it can permit production of the features without the need to combine the cell adhesive material with the agents prior to deposition on the coating or may provide more effective or concentrated contact between certain agents and cells than an approach in which agents are distributed throughout the feature, while the second approach may be advantageous in that it can permit production of the features using a single printing step or may allow a sustained release of agents from the features. In both cases, the coating comprising the cell adhesion resistant material resists adhesion of cells that are subsequently deposited onto the array, thereby substantially confining deposited cells (and their descendants, if any) to the cell adhesive regions.

[0071] In some embodiments, an array of the invention comprises a substrate having a coating comprising a cell adhesive material disposed on at least a portion of the surface of the substrate. The coating typically forms a layer that is relatively thin in height, as compared with its length and width and, in many embodiments, has a relatively uniform thickness. A cell adhesion resistant material is deposited on the coating in a pattern that leaves portions of the cell adhesive material exposed, creating a pattern of cell adhesive regions. In some embodiments, a mask is used to prevent deposition of the cell adhesion resistant material on those portions of the coating on which features are to be produced. Multiple features are produced by depositing agents atop the regions where the cell adhesive material is exposed. An agent may, but need not, be disposed atop all or a substantial portion of the region of cell adhesive material on which it is deposited. The cell adhesion resistant material resists adhesion of cells that are subsequently deposited onto the array, thereby substantially confining deposited cells (and their descendants, if any) to the cell adhesive regions.

[0072] In general, any suitable substrate having a surface that can support production of the features can be used to produce an array of the invention. The surface can be composed of a material that resists cell adhesion or can be rendered resistant to cell adhesion, e.g., by applying a cell adhesion resistant coating thereto. In various embodiments the surface can be composed at least in part of glass, plastic (such as polystyrene, polytetrafluoroethylene (PTFE), polyvinylidenedifluoride, poIy(ethylene terephthalate) (PET), polyurethane, polycarbonate, polypropylene, polymethyl methacrylate, natural or synthetic rubber, or a blend or copolymer of any of the foregoing), silicon, metal, (such as gold, stainless steel, or aluminum), polysaccharide (such as nitrocellulose, methylcellulose, cellulose, dextran, chitosan); mineral (such as quartz, graphite, or hydroxyapatite). It will be understood that many materials that are structurally related to certain of the foregoing materials exist and may be used in embodiments of the invention. For example, polymers having the same backbone but different side chains may be used. The substrate may be a non-porous solid or a porous solid in various embodiments. Often the material (or mixture thereof) is substantially transparent to light visible to the human eye, e.g., light with a wavelength in a range from about 380 nanometers (nm) to about 740 nm. The material may be substantially transparent to light over at least part of the UV range (about 10 nm to about 400 nm) and/or the infrared range (about 740 nm to about 300 micrometres).

[0073] In some embodiments, the surface is the bottom of a culture dish or culture chamber. In some embodiments, array is produced in a depression or recessed area of an article such as a slide. A cover can be placed above the depression or recessed area, which may be useful, e.g., to protect the array prior to cell seeding, during cell culture, during screening manipulations, or in order to facilitate preserving an array after performing a screen. In some embodiments a gas-permeable membrane (e.g., composed at least in part of polydimethylsiloxane (PDMS)) is used. In some embodiments an array is surrounded by a gasket or barrier to contain the culture medium. In some embodiments, the invention provides a culture vessel that has one or more depressions or recessed areas in its base into which an article comprising a surface can be inserted. "Culture vessel" refers to any container or receptacle that can be used to hold medium for culturing cells. Other components can be integrated with the surface. For example, optical or electrical components such as lenses, electrodes, sensors, etc., can be integrated with the surface. In some embodiments, the invention encompasses use of micro fluidics. For example, microchannels (e.g., having a diameter or cross-sectional dimensions on the submillimeter scale) can be provided between features, around the perimeter of the features, or under the features. Such microchannels may, for example, be used to deliver a test compound to an array.

[0074] In many embodiments, the surface (or at least the region of the surface in which the features are located) is substantially planar (flat). In some embodiments, the surface can have concave or convex regions. In some embodiments, the substrate comprises one or more detection elements, diffraction gratings, channels, or other elements. The scale of such elements can range from the micrometer to the nanometer in various embodiments. For example, the scale can be on the micron scale for microfluidic channels or on the nanometer scale for nanotubes or other nanoscale elements. Other components, such as lenses and electrodes, can be integrated with the surface. In some embodiments the surface can be the bottom of a culture dish or culture chamber or may be adapted for use with a particular culture dish or culture chamber. In general, the material of the substrate and geometry of the array can be selected based at least in part on criteria that it should be useful for automation of array formation, cell culturing, and/or detection of cellular phenotype.

[0075] Slides, e.g., microscope slides, provide a convenient surface and are available commercially in a range of sizes. Standard microscope slides with a surface of about 3" x 1" (75 millimeters (mm) x 25 mm) can be used. Larger or smaller surfaces could be used. For example, slides of about 4" x 3" (102 mm x 76mm), 4" x 3-1/4" (102 mm x 83 mm), 5" x 4" (127 mm x 102mm), 6" x 4-1/2" (152 x 1 14mm), 7" x 5" (178 mm x 127 mm) could be used. In some embodiments, the thickness is between about 0.5 mm and about 2.0 mm, e.g., between about 1.0 and about 1.4 mm, although thinner or thicker substrates can be used in various embodiments. Often the surface is rectangular or square, though other shapes such as circles or ovals, or polygonal shapes with 3, 4, 5, 6, or more sides could be used.

[0076] In many embodiments, an array comprises features arranged in rows and columns in which features in adjacent rows are located directly above one another so that the rows and columns form a grid of perpendicular lines (see, e.g., Supplementary Fig. 2(b), right panel). For example, an array can be an n x m array, where n is the number of rows and m is the number of columns, n and m can be the same or different. In some embodiments, an array comprises an "array of subarrays". For example, an array can comprise multiple individual n x m arrays (subarrays), where n and m can independently differ between different subarrays or may be the same. The subarrays themselves may be arranged in rows and columns, e.g., as &p x q array, wherein each element of the p q array is an n x m array of features. The subarrays are typically separated from each other by distances that are greater than the distance between features in the subarrays. In some embodiments, an array comprises an arrangement in which features in consecutive rows are offset from one another, e.g., by half the distance between features in the rows (see, e.g., Supplementary Fig. 1(a), left panel). [0077] In some embodiments, an array comprises at least 50 features per square centimeter (cm²), e.g., between 50 and 100 features/cm². In some embodiments, an array comprises between 100 and 1000 features/cm². In some embodiments, an array comprises between 1000 and 5000 features/cm² or between 5000 and 10,000 features/cm². In some embodiments, an array has at least 10,000 features/cm², e.g., up to 100,000 features/ cm², or up to 1,000,000 features/cm². In some embodiments the area of an array having, e.g., any of the foregoing feature densities, is at least 1 cm², e.g., between 1 cm² and 5 cm². In some embodiments, the area is between 5 cm and 10 cm², or between 10 cm² and 20 cm².

[0078] In some embodiments, an array comprises at least 90 features. In some embodiments, an array comprises at least 300 features. In some embodiments, an array comprises at least 1,000 features. In some embodiments, an array comprises between 1 ,000 and 3,000 features, or between 3,000 and 5,000, or between 5,000 and 10,000 features. In some embodiments, an array comprises between 10,000 and 20,000 features, or between 20,000 and 50,000, or between 50,000 and 100,000 features. In some embodiments an array comprises up to 100,000 features (e.g., any number of features up to 100,000. In some embodiments, an array comprises at least 100,000 features, e.g., between 100,000 and 1,000,000 features. In some embodiments the number of features is a multiple of 96, wherein the number may fall within any of the foregoing ranges.

[0079] In some embodiments, an array comprises at least 50 different features, wherein features differ from each other based on the agent(s) contained therein (e.g., the agents comprise different sequences of interest). In some embodiments an array comprises at least 90 different features. In some embodiments, an array comprises at least 300 different features. In some embodiments, an array comprises at least 1,000 different features. In some embodiments, an array comprises between 1 ,000 and 3,000 different features, or between 3,000 and 5,000, or between 5,000 and 10,000 different features. In some embodiments, an array comprises between 10,000 and 20,000 different features, or between 20,000 and 50,000, or between 50,000 and 100,000 different features. In some embodiments an array comprises up to 100,000 different features (e.g., any number up to 100,000. In some embodiments, an array comprises at least 100,000 different features, e.g., between 100,000 and 1,000,000 different features.

[0080] In some embodiments, an array includes replicates of at least some of the features. For example, an array may include replicates of between 1% and 100% of the features present thereon. The number of replicates can vary. For example, in some embodiments there are between 2 and 10 replicates of at least some features. In some embodiments there are between 3 and 6 replicates of at least some of the features. Replicates may be positioned in any way. For example, replicates may be adjacent to one another or clustered together or may be dispersed throughout the array.

[0081] In some embodiments, an array comprises one or more "control features". The control features can comprise an agent whose expected effect on diverse cells is known. Different control features may comprise different agents. The control features may be used, for example, to verify that infection, transfection, selection, or induction of expression occurred effectively. An exemplary control agent may be, for example, a vector comprising an expression construct that encodes a fluorescent protein. In some embodiments an array comprises one or more cell adhesive regions of the same shape and size as the features but not comprising an agent. Such region(s) may have a printing buffer (e.g., of the same composition as that used to print the agents) deposited thereon. Such regions may serve as control regions and can be used, for example, to verify that a printing buffer is free of agents, to quantify interfeature migration of cells (e.g., in order to verify a lack of interfeature migration), etc.

[0082] In some aspects, the invention provides arrays having a variety of different shapes, dimensions, interfeature distances, and configurations. In some nonlimiting embodiments, the features are circular and have a diameter of about 100 μπι - about 1,000 μηι. For example, in some embodiments the features have a diameter of about 100 μιη - about 200 μιτι, about 200 μηι - about 300 μιτι, or about 300 μηι - about 600 μιτι. In some nonlimiting embodiments, the features have sides or axes about 100 μπι - about 1,000 μιη long. For example, in some embodiments the features have sides or axes about 100 μηι - about 200 μιη, about 200 μηι - about 300 μηι, or about 300 μηι - about 600 μιη long. In some embodiments, the interfeature distance is between about 25 μιη and 300 μιη.

"Interfeature distance" refers to the minimum distance between adjacent features, i.e., the distance between the points where the borders of adjacent features are closest to one another. In some embodiments, the interfeature distance is between about 50 μηι - about 100 μηι, or between about 100 μηι - about 200 μιη. In some embodiments features are circular or square and have a center-to-center distance of between 500 μηι and 1 mm, e.g., about 750 μηι. In some embodiments, features are approximately 600 μιη in diameter with 750 μπι center- center spacing. [0083] In some embodiments, the distance between edges of adjacent features is between about 20 μΜ and about 100 μΜ. In some embodiments, the distance between edges of adjacent features is between about 20 μ and about 50 μΜ. In some embodiments, the distance between edges of adjacent features is between about 50 μΜ and about 100 μΜ. In some embodiments, the distance between edges of adjacent features is about 20 μΜ, about 25 μΜ, about 30 μΜ, about 35 μΜ, about 40 μΜ, 45 μΜ, and about 50 μΜ.

[0084] In some embodiments, features can be formed at least in part within depressions or cavities of a surface. The depressions or cavities typically have a diameter, axis, or longest side about 50 μηι - 1 mm in length, e.g., 300 μπι - 600 μτη, and may be referred to as microwells. Typically the depth of the microwells is between about 50 μιη - 500 μιη. The depressions or cavities help to confine the cells to the region of the features. A surface patterned with depressions or cavities can be produced using, e.g., soft lithography, engraving, etching, or any other suitable method. In some embodiments, the depressions have sloping walls. In some embodiments the walls are perpendicular to the surface. In some embodiments, depressions or cavities are formed in a flexible or moldable material such as PDMS. For example, a slab comprising microwells can be placed on the surface of a support such as a slide. In some embodiments the slab contains multiple holes extending from the top surface to the bottom surface of the slab. When placed on top of a surface, these holes form microwells whose bottom is formed by the surface on which the slab is placed. In some embodiments the microwell sides and bottom are formed by the slab material. See PCT/US2006/036282 (WO/2007/035633) -SCREENING ASSAYS AND METHODS for discussion of construction of an article comprising multiple microwells. The use of depressions or cavities to confine cells to the features may be used in combination with any of the other approaches described herein in various embodiments of the invention.

[0085] In some embodiments, topographic variation on a micron or submicron scale (also termed "microtexturing") is used to promote cell adhesion or to inhibit cell adhesion or migration. "Topographic variation" in this context refers to deviation from a substantially planar surface. For example, upwardly projecting three-dimensional structures such as dots or pillars or downwardly projecting grooves or pits, constitute topographic variation. In some embodiments, such structures have dimensions (e.g, height, depths, width, diameter) of 1 μπι - 10 μηι. In some embodiments, such structures have dimensions (e.g., height, depths, width, diameter) of less than 1 μτη, e.g., 1 nm - 100 μπι. Texturing of a surface can be altered by additive methods, such as by depositing particles, applying a coating layer comprising dots, pillars, or other particles, or in situ synthesis, or by removing some of the surface by subtractive methods such as mechanical removal or chemical etching. The use of topographic variation to confine cells to the features may be used in combination with any of the other approaches described herein in various embodiments of the invention.

[0086] In some aspects, an array of the invention can be stably stored at -80°C for at least 8 months without substantial loss of activity. For example, the percentage of features that yield stable infection or transfection of cells can be at least 80%, 90%, 95%, 98%, or more as compared with a newly produced array. Whether cells are stably inlected or translected can be tested, for example, by examining features that comprise an agent that confers drug resistance on cells and determining the percentage of such features that yield drug-resistant cell spots.

[0087] Cell Adhesion Resistant Materials

[0088] Any suitable cell adhesion resistant material can be used in various embodiments of the invention. In many embodiments, the material is a biocompatible material that does not substantially dissolve when exposed to an aqueous environment (such as tissue culture medium) in the form in which it is present in an array of the invention. In general, a cell adhesion resistant material may be one that resists nonspecific adsorption of proteins. In some embodiments, a material useful to minimize protein adsorption in applications such as implantable or indwelling medical devices, diagnostic devices that may come in contact with protein-containing body fluids, or DNA or protein microarrays that seek to detect affinity- based interactions between a surface-bound probe (e.g., a nucleic acid or antibody) and a target molecule in a sample, may be used. Such materials or surfaces comprising them are sometimes referred to as "non-fouling". In some embodiments, a hydrophilic material is used. "Hydrophilic" in this context refers to materials that interact readily with water, typically via hydrogen bonds. Such materials often contain polar or charged groups. A hydrophilic material in dry form may readily adsorb or take up water. Without wishing to be bound by any theory, protein adsorption onto hydrated, hydrophilic surfaces may not be energetically favorable, which may inhibit cell adhesion. In some embodiments, a zwitterionic polymer is used as a cell adhesion resistant material.

[0089] Exemplary cell adhesion resistant substances include, e.g., a variety of polypeptide or non-polypeptide polymers such as acrylamides (e.g., polyacrylamide, polymethacrylamide), acrylates (e.g., poly (hydroxyethylmethacrylate) (pHEMA), polymers of ethylene oxide (poly(ethylene glycol) (PEG)), polysaccharides such as heparin or dextran, polypeptides such as albumin, and copolymers, mixtures, blends, and derivatives of any of the foregoing. In some embodiments a polymer is a comb polymer or a brush polymer. As used herein, a "derivative" of a polymer includes, for example, graft polymers or comb polymers that have the polymer as a backbone or as a side chain, A "brush polymer" refers to an assembly of polymer chains attached by one end to a surface. The chains are typically dense enough so that there is crowding among the polymer chains, which forces them to extend away from the surface to avoid overlapping. In some embodiments a copolymer comprises 2 or 3 different monomers. The ratio of monomers can vary.

[0090] In some embodiments, polyacrylamide is used as a cell adhesion resistant material. In some embodiments, a copolymer of acrylamide and one or more other chemical species, such as an acrylic acid or a salt thereof, can be used. In some embodiments, a polymer comprises at least 10% acrylate subunits. In some embodiments a copolymer comprising acrylate and methacrylate or 2-hydroxyethyl me hacrylate monomers is used. In some embodiments, a polyacrylamide-based polymer having a modified side chain, such as poly(iV-isopropylacrylamide) is used.

[0091] In some embodiments, a cell adhesion resistant material comprises PEG (a polymer of ethylene oxide subunits) or a derivative thereof or a colymer comprising at least some ethylene oxide subunits. A wide variety of PEGs and related polymers are known in the art and may be used in various embodiments.

[0092] In some embodiments, a polymer with a comb-like architecture, comprising a relatively hydrophobic backbone (such as a polymethacrylate) and hydrophilic side-chains (such as a PEG) is used as a cell adhesion resistant material.

[0093] In many embodiments, a cell adhesion resistant material is present on an array surface as a hydrogel, e.g., as a hydrogel layer atop a rigid support such as a glass or plastic slide or culture dish. A "hydrogel" may be defined as a two- or multicomponent system comprising a three-dimensional network of polymer chains and water that fills the space between them. The polymer chains are relatively hydrophilic and, in many embodiments, the polymer is water-soluble when in non-crosslinked form. A hydrogel may be in the form of a colloidal gel in which water is the dispersion medium. Materials useful for forming hydrogels include, e.g., polyacrylamide, poly(ethylene glycol), polyvinyl alcohol), polyvinylpyrrolidone, poly(hydroxyethyl methacrylate), or various polysaccharides. In general, any of a variety of different polymeric materials can be used to form hydrogels in various embodiments of the invention. In most embodiments, the hydrogel is formed from a material that does not comprise a significant number of integrin ligands (e.g., RGD peptides) or other motifs that might promote cell adhesion. In many embodiments, the polymer is not a polypeptide. In some embodiments, a hydrogel contains over 90% water, e.g., at least 95%, or at least 99% water. Gelation can be achieved, e.g., by physical, ionic, or covalent interactions. In some embodiments a hydrogel layer is about 5-30 μπι in thickness, e.g., about 10 μιη thick. In some embodiments a thinner or thicker layer is used. For example, in some embodiments a hydrogel has a thickness of about 10 - 100 nm. It will be understood that a hydrogel-forming material can be dried and will typically be thinner in dried form than when hydrated.

[0094] In some embodiments, polystyrene or glass that has not been treated to render it suitable for tissue culture is used as a cell adhesion resistant material.

[0095] In general, a polymer can be synthesized using any of various polymerization techniques known in the art, such as cationic, anionic, radical, ring-opening metathesis, photochemical, or electrochemical polymerization. One of ordinary skill in the art will select an appropriate method depending, e.g., on the particular monomers to be used. In various embodiments a polymer may have an average molecular weight (arithmetic mean) of between about 200 Da and about 100- 200 kDa, e.g., about 300; 500; 1,000; 1 ,500; 2,000; 5,000; 10,000; 20,000; 30,000; 40,000; 50,000; 60,000; 70,000; 80,000; 90,000; orl 00,000 daltons.

[0096] In some embodiments, a polymer comprises intermolecular or intramolecular cross-links. In general, cross-linking can be via covalent or non-covalent bonds. Cross-links can be formed by chemical reactions that may, for example, be initiated and/or promoted by heat, pressure, change in pH, radiation, UV light, other physical means, and/or use of chemical cross-linking agents. One of ordinary skill in the art will be able to select appropriate cross-linking agents. For example, polyacrylamide can be cross-linked using, e.g., Ν,Ν'-methylenebisacrylamide or 2-hydroxyethyl methacrylate. The nature and density of cross-linking can be controlled as known in the art and may be selected based at least in part on desired properties of the resulting polymer. For example, high cross-linking densities can result in a material with greater strength and/or rigidity.

[0097] In some embodiments a polymer is attached, e.g., covalently attached, to a surface. The polymer may comprise a reactive functional group, e.g., at either or both ends of the chain, which may be used to attach the polymer chain to a surface. In some embodiments a polymer is crosslinked to itself and is crosslinked or otherwise attached to a surface. A surface (e.g., glass) may be treated with a compound that comprises functional groups suitable to crosslink or covalently bind to the polymer. In some embodiments a substrate (e.g., glass or other substrate having exposed hydroxyl groups) is silanized (treated with alkoxysilane molecules such as aminosilanes), which may comprise functional groups suitable for polymerization or for reaction with a polymer.

[0098] In some embodiments a commercially available hydrogel-coated substrate, e.g., a hydrogel-coated slide, is used. Slides that are suitable for use in nucleic acid or protein microrrays may be used in various embodiments. Examples include, e.g., CodeLink® (Surmodics), Hydrogel PE or HydroGel (Perkin Elmer). Other examples include, e.g., Slide H (Schott Nexterion) comprising OptiChem coating (Accelr8). In some embodiments a hydrogel comprises amine-reactive functional groups, which may be deactivated prior to use of an array. In some embodiments, a polyacrylamide hydrogel-coated surface is made as described in Flaim, CJ, et al., Nat Methods. (2005), 2(2): 119-25, or references therein.

[0099] Cell Adhesive Materials

[00100] Any suitable cell adhesive material can be used in various embodiments of the invention. In many embodiments, the material is a biocompatible material that does not substantially dissolve when exposed to an aqueous environment (such as tissue culture medium) in the form in which it is present in an array of the invention. In some embodiments a cell adhesive material may promote cell adhesion by, for example, interacting with a cell surface molecule, e.g., a cell adhesion molecule such as an integrin. In some embodiments, a cell adhesive material comprises one or more extracellular matrix (ECM) components or a portion or mimetic thereof. ECM components include a variety of proteins and glycosaminoglycans (GAGs). Protein components of the ECM include collagens (e.g., collagen I, III, or IV), elastin, fibronectin, laminin, and vitronectin. In some embodiments, the cell adhesive material comprises an integrin ligand. As known in the art, integrin ligands frequently contain the tripeptide arginine-glycine-aspartic acid (RGD), which interacts with integrins. In some embodiments the cell adhesive material has a surface density of integrin ligands of at least 1 ng/cm². In some embodiments, a hydrophobic and/or positively charged material is used as a cell adhesive material. In some embodiments a hydrophobic material is a nonpolar material that lacks the capacity for hydrogen bond formation with water.

[00101] In some embodiments, a gelatin is used as a cell adhesive material. In some embodiments, "gelatin" is a heterogeneous mixture of water-soluble proteins produced by partial hydrolysis of proteins, e.g., collagen, extracted by boiling tissues such as skin, tendons, ligaments, bones, etc. in an aqueous medium, typically water, as known in the art. Type A gelatin is derived from acid-treated tissue and Type B gelatin is derived from alkali- treated (e.g., lime-treated) tissue. In some embodiments, a type B gelatin is used as a cell adhesive material. In some embodiments, a gelatin is bovine derived. For example, the gelatin may be derived from bovine skin. In some embodiments, a gelatin is porcine or equine derived. In some embodiments, a gelatin is derived from fish, sponges, or other vertebrate or invertebrate organisms. In some embodiments a recombinant gelatin is used. For example, recombinant collagen can be expressed, purified, and denatured (with or without chain fragmentation) to yield recombinant gelatin. In some embodiments, a specified fragment of a selected collagen a chain may be produced and denatured. See, e.g. Olson, D., Advanced Drug Delivery Reviews 55 (2003) 1547- 1567, for discussion of recombinant gelatin and collagen and methods of making thereof.

[00102] In some embodiments, an at least partly purified ECM component or mixture thereof is used. In some embodiments a recombinant ECM component or mixture thereof is used. A variety of commercially available ECM-derived matrices can be used in various embodiments of the invention. Examples include Matrigel™ (Becton Dickinson) from the Engelbreth-Holm-Swarm (EHS) sarcoma, EHS Natrix (BD Biosciences), ECL (US Biological).

[00103] In some embodiments, a molecule that is not naturally cell adhesive is rendered cell adhesive by covalently or noncovalently attaching an integrin ligand thereto. For example, an RGD-containing peptide can be attached to various synthetic or naturally occurring polymers either prior to or after deposition on an array surface. Any of a variety of methods can be used to attach a peptide (or other molecue) to a material. One of ordinary skill in the art will be able to select an appropriate method based, e.g., on factors such as available functional groups of the material. See, e.g., Hermanson, G., Bioconj gate Techniques, 2^nd ed., Academic Press (2007) and The Molecular Probes Handbook, supra. The RGD-containing peptide may comprise a lysine or cysteine residue to provide an amine or thiol functional group, respectively, which can react to form a covalent bond with an appropriate reactive functional group in the molecule to be modified, or with a bifunctional crosslinker capable of reacting with a functional group of the molecule to be modified. Amine-reactive groups include, e.g., succinimidyl esters and sulfonyl chlorides. Thiol- reactive groups include iodoacetamides, maleimides, and benzylic halides. [00104] In some embodiments, a polycationic polypeptide, e.g,. a polyaminoacid such as polylysine (e.g., poly-L-lysine) or polyomithine (e.g., poly-L-ornithine) is used.

[00105] Any of a range of different concentrations of a cell adhesive material can be present in a composition to be deposited, in various embodiments. For example, the concentration can be about 0.1% to about 0.5% (or about 1 mg/ml to about 5 mg/ml expressed in terms of weight/volume) in certain embodiments. In some embodiments the concentration is about 0.2% (2 mg/ml). For example, 0.2% gelatin is used in some embodiments. In general, an appropriate concentration for any particular cell adhesive material can be empirically determined. The cell adhesive material can be dissolved in a suitable vehicle. For example, deionized or distilled water can be used.

[00106] In some embodiments a composition comprising a cell adhesive material (e.g., gelatin) contains a biocompatible compound that serves as a humectant or surfactant (which term refers to compounds that lower the surface tension of a liquid, the interfacial tension between two liquids, or that between a liquid and a solid). The compound may, for example, reduce pin fouling or improve spreading properties of the composition. In some embodiments, a polyol such as glycerol is used. For example, a polyol such as glycerol can be present at, e.g., a concentration of about 0.05% to about 0.2%, e.g., about 0.1%.

[00107] A cell-adhesive material is deposited in a plurality of discrete locations on the surface that resists cell adhesion, resulting in a surface that contains multiple discrete cell- adhesive regions separated by areas that are cell-resistant. Typically, the cell-adhesive regions are arranged in a regular format such as in rows and columns, which are optionally perpendicular to each other. The size, shape, number, and spacing of the regions can vary. For example, the perimeters of the regions can be circles, rectangles, ovals, etc. In some embodiments, the regions are completely or substantially completely covered with the cell- adhesive material. A variety of cell-adhesive materials can be used. In some embodiments, a cell-adhesive material is a hydrogel. In some embodiments, the cell-adhesive material comprises one or more proteins, e.g., a protein found in the extracellar matrix (ECM). Examples of suitable proteins include, e.g., collagens (e.g., collagen I, III, IV, etc.), laminin, fibronectin, vitronectin, and mixtures thereof. In another embodiment, polylysine is used. In some embodiments, the material comprises a protein that has been subjected to one or more processing steps that modifies its physical and/or chemical structure. For example, a protein may be denatured and/or partially hydrolyzed. In some embodiments, the cell-adhesive material is a gelatin, e.g., a material produced by partial hydrolysis of collagen. Other synthetic and naturally occurring cell adhesive materials could be used. A cell-adhesive material could be obtained from natural sources, produced recombinantly, chemically synthesized, or produced using any other suitable method.

[00108] The methods localize cell adhesion and infection only to the areas on the microarray surface that contain the cell-adhesive material.

[0 10 1 Assessina Cell Adhesion or Migration

(00110] If desired, cell adhesion or cell migration to or on a material of interest may be assessed using any suitable method. In some embodiments, cell adhesion or cell migration is assessed in the context of an array. In some embodiments, a surface is printed with a pattern of materials to be assessed (e.g., as for an array), but without agents affixed thereto. Cell adhesion to a material of interest can be measured by plating cells of an adherent cell line onto a surface (e.g., the surface of a tissue culture dish or slide) composed of or coated with the material, maintaining the surface under conditions suitable for cell adhesion for a selected time period to allow the cells to settle on the surface, and measuring the number of adherent cells on a selected area (or areas) of the surface after such time period. Typically the surface is washed gently prior to measuring the number of adherent cells. The selected time period is typically at least 15 minutes and may range up to about 12 hours, or more. Often it is kept sufficiently short that significant cell proliferation does not occur. Cell adhesion to multiple distinct areas (e.g., 3-6 areas) can be measured and the results averaged. Results can be expressed, for example, as the number of adhering cells/cm².

[00111 ] The number of adherent cells within an area of interest can be assessed using any of a number of techniques in various embodiments of the invention. For example, cells can be detected using microscopy, e.g., phase contrast microscopy or differential interference contrast microscopy. Fluorescence microscopy can be used if, e.g., the cells express a fluorescent protein or are stained with a fluorescent substance. In some embodiments cells are labeled prior to or following plating to facilitate their quantification. In general, cells can be labeled using any approach known in the art that can facilitate detection of cells or cellular biomolecules or structures. In some embodiments, measuring the number of cells could comprise removing the cells located on a selected area of the surface (which in some embodiments may be the entire surface to which cells could come in contact and potentially adhere) and measuring the number of cells removed. In some embodiments, cells are removed from the surface by enzymatic means (e.g., trypsinization) or by mechanical means, [00112] Any of a variety of devices suitable for counting cells can be used, such as a hcmocytometer, electronic particle counter (e.g., a resistance-based counter, examples of which include the Beckman Coulter Zl and Z2 and the Innovatis CASY) that measures cell number based on the change in current generated when a cell passes through a narrow orifice, (b) image analysis of a microscope view of unstained or stained cells or nuclei in special counting chambers by, e.g., visible light or fluorescence (examples of instruments useful for this purposes include, e.g., the Countess (Invitrogen), Cellometer (Peqlab), and

Nucleocounter (Chemometie), or a flow cytometer. In many embodiments an automated cell counting method is used (i.e., a method that does not require an individual to count cells by eye with a microscope). As described herein, a standard fluorescence seamier can be used to quantify cell number if cells are appropriately labeled. In some embodimenlscells that have adhered to or migrated a surface (or portion thereof) are quantified indirectly, i.e., without counting individual cells or nuclei. For example, in some embodiments, cell number is measured by measuring a total amount of a biomolecule such as DNA or protein that can serve as an indicator of cell num ber.

1 113] DNA can be stained with DAPI, propidium iodide, a cyanine dye (such as PicoGreen, SYBR Green, SYBR Gold), or a bis-benzimide dye (such as Hoechst 33258 and Hoechst 33342) and subsequently detected based on fluorescence. Cells can be stained for a specific protein such as actin, or total protein can be measured. In some embodiments, activity of a cellular enzyme is used as an indicator of cell number. For example, an MTT or MTS assay can be used. As known in the art, MTT assays are colorimetric assays that measure the activity of enzymes that reduce MTT or similar dyes (e.g., XTT, MTS, water- soluble tetrazolium salts (WSTs) to formazan dyes, giving a purple color.

[00114) It will be understood that in many embodiments useful information regarding the ability of a material to resist or promote cell adhesion or migration can be obtained without an absolute cell count. For example, when comparing the properties of different materials, relative measurements may be used. If desired, appropriate standard curves can be generated that can permit a comparison between results obtained using different assays or to permit determination of an absolute cell number.

|00115| Approaches similar to those useful for measuring cell adhesion can be used to measure cell migration on a material of interest. In general, such methods can comprise providing a first surface having adherent cells attached thereto, providing a second surface comprising a material of interest adjacent to the first surface, and measuring the number of cells that migrate from the first surface onto the second surface within a selected time period, hi some embodiments, cells are deposited onto one or more defined region(s) of a surface, and the number of cells that have migrated away from the region is measured after a selected time period. In some embodiments time-lapse video microscopy can be used. In some embodiments, a "scratch assay", also termed a "wound healing assay" may be used to measure cell migration (see, e.g., Lampugnani MG: Cell migration into a wounded area in vitro. Methods Mol Biol (1999) 96: 177-182). The assay can comprise providing a surface having a confluent monolayer of cultured cells thereon, producing a "wound" in the monolayer, and measuring migration of cells into the area of the wound. The monolayer fills in ("heals") the wound in a process that can typically be observed over a time period of about 3-24 hours, during which time cells typically polarize toward the wound, initiate protrusion, migrate, and close the wound. The extent of migration, e.g., the position of the front edge of the migrating cell monolayer, during a selected time period (e.g., 5-10 hours) can be measured. The "wound" can be made by, e.g., dragging an implement such as a pipette tip, ruler, pin, etc., across the cells, resulting in scratches with a width of, for example, about 0.1 to about 0.5 mm.

[00116] In some embodiments, cell number, cell adhesion, cell migration, or cell phenotype is assessed at least in part using imaging software such as CellProfiler™ cell image analysis software, available at http://www.cellprofiler.org/ (Carpenter, A., et al., Genome Biology 2006, 7:R100), MetaMorph® Microscopy Automation & Image Analysis Software (Molecular Devices, LLC), ImageJ (Rasband, W.S., ImageJ, U. S. National Institutes of Health, Bethesda, Maryland, USA, http://imagej.nih.gov/ij/, 1997-201 1 ;

Abramoff, M.D., et al., Biophotonics International, 11(7): 36-42, 2004.), etc. In some embodiments, cell number, cell migration, or cell phenotype is assessed at least in part using microarray image analysis software such as GenePix® software (Molecular Devices).

[00117] It will be understood that an array (or a region of an array) or a material that is considered to have a particular property of interest with respect to mammalian cells (e.g., an array that comprises means effective to substantially confine adherent mammalian cells to the features for a period of time, or a material that inhibits mammalian cell adhesion or migration or a material that promotes mammalian cell adhesion) need not and often may not possess that property with respect to all available adherent mammalian cell types or cell lines. For purposes of the present invention, an array (or region thereof) or a material will be considered to have a particular property of interest if it exhibits that property when tested with at least one set of diverse adherent mammalian cell lines, "Diverse adherent mammalian cell lines" typically refers to a set of at least 5 distinct adherent mammalian cell lines, e.g., at least 10 distinct adherent mammalian cell lines, wherein the cell lines are derived from different tissues of origin. In some embodiments, a set includes cell lines derived from multiple different species (e.g., human cell lines and mouse cell lines). In some embodiments, a set of diverse adherent mammalian cell lines consists of cells derived from a single species (e.g., humans). In some embodiments, a set of diverse adherent mammalian cell lines includes cell lines derived from at least 5 different tissues. In some embodiments, a set of diverse adherent mammalian cell lines includes at least one cell line having a fibroblast type and at least one cell line having an epithelial type and, in some embodiments, at least one cell line having a neuronal, glial, endothelial, hepatocyte, melanocyte, macrophage, or keratinocyte type. In some embodiments, a set of diverse adherent mammalian cell lines comprises at least one cell line of at least 2, 3, 4, 5, 6, or all of the foregoing cell types. In some embodiments, a set of diverse mammalian cell lines comprises at least 5, 10, 12, or 15 cell lines listed in

Supplementary Table 2. In some embodiments, a set of diverse mammalian cell lines includes at least one highly migratory cell line, such as 786-0 cells, U87 cells, or 90-8T cells. In some embodiments an inventive array is effective to substantially confine 786-0 cells, U87 cells, and/or 90-8T cells to the features. In some embodiments, an array or region or material would exhibit a property of interest when tested with most (e.g., at least 50%, 60%, 70%, 80%, 90%, 95%, or more in various embodiments) distinct mammalian cell lines randomly selected from a collection of cell lines held by a cell bank or repository.

[00118] In some embodiments, an array meets the following criteria with regard to a diverse set of mammalian cell lines: (a) features containing an agent yield surviving cells following 2-5 day drug selection to remove uninfected cells, while control regions containing no agent yield essentially no surviving cells following the same treatment; (b) cells are surviving and, in some embodiments, proliferating, on the features for long enough to do an assay of interest (e.g., 3-7 days); (c) cells are attached well enough that arrays can be gently manipulated without the cells falling off; and (d) cells appear to grow normally and assume an appearance that is not markedly different (e.g., is reasonably similar) to the way they appear when plated on standard tissue culture plastic.

[00119] III. Agents

[00120] Any of a variety of different agents may be used in an array of the invention, in various embodiments. In some embodiments, an agent comprises a nucleic acid to be introduced into eukaryotic cells. In some embodiments, an agent is a virus capable of infecting mammalian cells and introducing a nucleic acid into said cells. In some embodiments of interest, an agent is a lentivirus capable of infecting mammalian cells and introducing a nucleic acid into said cells. In some embodiments, a nucleic acid comprises an expression cassette that directs transcription of an RNA (e.g., an shRNA or an mRNA) within the cell. Following such transcription the effect of expressing the RNA in the cell can be assessed. For example, expressing a shRNA allows assessment of the effect on cell phenotype of inhibiting expression of the gene(s) that are targets of the shRNA. Expressing an open reading frame (ORF) that encodes a protein allows assessment of the effect of the protein on cell phenotype. In some aspects, the invention provides arrays useful for reverse transfection or reverse infection of agents into eukaryotic cells. In some aspects, the invention provides improvements in reverse transfection/reverse transfection arrays, methods of making thereof, and/or methods of use thereof.

100121] In some embodiments, an array comprises a library of agents that represents a set of genes. An agent will be said to "'represent a gene" if the agent can be matched to a particular gene based on sequence of the gene or sequence or structure of a gene product encoded by the gene. In some embodiments, a nucleic acid represents a gene if the nucleic acid comprises a sequence that is substantially or perfectly identical or substantially or perfectly complementary to at least a portion of the gene over a continuous sequence of at least 10, 50, 100, 500 nucleotides, or more. One of ordinary skill in the art would readily be able to obtain nucleic acid sequences of genes and portions thereof, e.g., nucleic acids sequences encoding RNAs or proteins, from publicly available databases such as those available at the National Center for Biotechnology Information (www.ncbi.nih.gov).

Exemplary databases include, e.g., GenBank, RefSeq, Gene, UniProt, SwissProt, and the like.

[00122] In some embodiments a set of genes comprises at least about 50%, 60%, 70%, 80%, 90%, 95%, or more of the genes present in a genome of a species of interest. In some embodiments the genome of the species of interest has been sequenced. For purposes of the present invention the number of genes in a genome will be taken to be the number of predicted genes in a sequence annotated as of the filing date of the present application. Sequenced genomes from a number of different organisms are available in publicly available databases such as the "Genome" database of the National Center for Biotechnology Information (http://www.ncbi.nlm.nih.gov/sites/Kenome). Other databases devoted to particular organisms are also available. In some embodiments the library represents at least 10,000 genes, e.g., at least 15,000, or at least 20,000 genes. In some embodiments the genes are human genes. In some embodiments the genes are mouse genes.

[00123] In some embodiments, an open reading frame (ORF) or short hairpin RNA (shRNA) library is used. A number of such libraries are known in the art. (See, e.g., references 3, 6, 1 0, 1 1). For example, a number of lentiviral (ORF and shRNA libraries are available, wherein expression of an ORF or shRNA in mammalian cells is achieved by infecting such cells with lentivirus comprising an expression cassette appropriate to direct such expression.

[00124] In some instances, a gene can encode multiple different gene products. The gene products may, for example, differ as a result of alternative splicing, RNA editing, etc., which may be tissue-specific or otherwise conditional. In some embodiments an array comprises features corresponding to multiple gene products for one or more such gene(s), e.g., 2, 3, or more gene products. In some embodiments a single gene product is represented on the array. For example, the gene product that is most widely expressed across tissue types or in a disease or tissue of interest may be selected. In some embodiments multiple distinct alleles of a gene are present in a population. For example, single nucleotide polymorphisms (SNPs) are common. Such SNPs may or may not alter a coding sequence (which may or may not alter the sequence of an encoded protein), regulatory sequence, intron, or other portion of a gene or intergenic region. In some embodiments, an agent corresponds to multiple different forms. For example, an RNAi agent may target a nonpolymorphic region or may be effective to inhibit multiple different polymorphic forms. In some embodiments an agent corresponds specifically to a particular polymorphic form. In some embodiments, multiple agents that correspond to different polymorphic forms of a gene or gene product are present on an array. In some embodiments an agent corresponds to a reference sequence present in the NCBI Reference Sequence (RefSeq) database (http://wvvw.ncbi.nlm.nih.gov/RefSeq/^'). In some embodiments, an agent represents the most common allele at a particular polymorphic position. Examples of polymorphic variants can be found in, e.g., the Single Nucleotide Polymorphism Database (dbSNP) (available at the NCBI website at

www.ncbi.nlm.nih.gov/projects/SNP/), which contains single nucleotide polymorphisms (SNPs) as well as other types of variations (see, e.g., Sherry ST, et al. (2001). "dbSNP: the NCBI database of genetic variation". Nucleic Acids Res. 29 (1): 308-31 1 ; Kitts A, and Sherry S, (2009). The single nucleotide polymorphism database (dbSNP) of nucleotide sequence variation in The NCBI Handbook [Internet]. McEntyre J, Ostell J, editors. Bethesda (MD): National Center for Biotechnology Information (US); 2002

(www.ncbi.nlm. nih.gov/bookshelf/br.fcgi?book=handbook&part=ch5). In some embodiments, multiple agents that provide different expression levels of a gene could be included on an array. Varying expression levels could be achieved, for example, by expressing a nucleic acid of interest under control of promoters of different strengths, different numbers of copies, or by expressing RNAi agents that result in different levels of inhibition of a gene of interest.

I (10125J In some embodiments a set of genes into a category of interest. For example, in some embodiments a set of genes shares a common biological function or activity or subcellular localization (e.g., the gene products of such genes share a common biological function or activity), chromosomal location, or regulation. It will be understood that the term "common" in this context means similar, highly similar, closely related, identical, or substantially identical. For example, the set of genes may comprise enzymes that have a common activity or act on a common substrate, DNA or RNA binding proteins, transcription factors, transcriptional co-activators or repressors, epigenetic modifiers (e.g., histone modifying enzymes, DNA modifying enzymes), cell cycle control proteins, etc. In some embodiments, an enzyme is a kinase, phosphatase, ATPase, GTPase, protease, ubiquitin ligase, deubiquitinase, acetylase, deacetylase, methylase, demethylase, acyltransferase, cytochrome P450 (CYP) family member, etc. In some embodiments, a set of genes encodes receptors. For example, the genes may encode G protein coupled receptors (GPCRs), nuclear hormone receptors, cytokine receptors, growth factor receptors, etc. In some embodiments a set of genes comprises at least 10 distinct members, e.g., between 10, 20, 50, 100, 200 and up to about 500 distinct members. In some embodiments a category of interest comprises genes that are involved in a particular biological process of interest, such as the cell cycle, apoptosis, autophagy, differentiation, endocytosis, membrane transport, protein quality control, protein degradation, signal transduction, transcription, translation, etc. In some embodiments a category of interest comprises secreted proteins, transmembrane proteins, mitochondrial proteins, or nuclear proteins.

[00126j i some embodiments a set comprises at least 80%, 90%, 95%, or more of the known or predicted members of a category of interest encoded by the genome of a species of interest. "Known" in this context refers to having been experimentally validated using art- accepted techniques or otherwise generally accepted in the art. "Predicted" in this context means that a member is reasonably expected to fall into a particular category based on art- accepted information pertaining to such member and using art-accepted prediction methods but has not necessarily been experimentally validated. For purposes hereof, "predicted" can refer to a prediction based at least in part on presence of particular sequence motifs, domains, or homology. In some embodiments, "predicted" function or activity is a prediction contained in a public database such as those available at the NCBI or UniProt or Gene Ontology. In some embodiments a category of interest is a protein family, wherein the proteins are related by sequence. For example, in some embodiments members of the family may be at least 50%, 60%, 70%, 80% identical, or more across a continuous sequence of at least 100, 200, 300, 500, or 1,000 amino acids.

|00127] In certain embodiments kinases are of interest. Kinases play important roles in diverse cellular and developmental processes including cell cycle progression, metabolism, and angiogenesis, among others, and are key components of numerous signal transduction pathways. Inhibition of kinases as a therapeutic strategy is of considerable importance. For example, kinase inhibitors have been approved for use in treating a variety of different cancers and show promise in a number of non-oncologic indications. Kinases can be classified based on the nature of their typical substrates and include protein kinases (i.e., kinases that transfer phosphate to one or more protein(s)), lipid kinases (i.e., kinases that transfer a phosphate group to one or more lipid(s)), nucleotide kinases, etc. Protein kinases (PKs) are of particular interest in certain aspects of the invention. PKs are often referred to as serine/threonine kinases (S/TKs) or tyrosine kinases (TKs) based on their substrate preference. Serine/threonine kinases (EC 2.7.1 1.1) phosphorylate serine and/or threonine residues while TKs (EC 2.7.10.1 and EC 2.7.10.2) phosphorylate tyrosine residues. Λ number of "dual specificity" kinases (EC 2.7.12.1 ) that are capable of phosphorylating both serine/threonine and tyrosine residues are known. The human protein kinase family can be further divided based on sequence/structural similarity into the following groups: (1) AGC kinases - containing PKA, PKC and PKG; (2) CaM kinases - containing the

calcium/calmoduiin-dependent protein kinases; (3) CK1 - containing the casein kinase 1 group; (4) CMGC - containing CDK, MAPK, GSK3 and CLK kinases; (5) STE - containing the homologs of yeast Sterile 7, Sterile 1.1 , and Sterile 20 kinases; (6) TK - containing the tyrosine kinases; (7) TKL - containing the tyrosine-kinase like group of kinases. A further group referred to as "atypical protein kinases" contains proteins that lack sequence homology to the other groups but are known or predicted to have kinase activity, and in some instances are predicted to have a similar structural fold to typical kinases. See, e.g.. Matthews, DJ and Gerritson, M., Targeting Protein Kinases for Cancer Therapy, Wiley, 2010, for further information regarding kinases.

[00128J Tn some embodiments, a set comprises variant kinases, e.g., mutant kinases or fusion protein kinases, wherein the mutation or fusion contributes to kinase inhibitor resistance or is an activating mutation that causes or contributes to a disease, e.g., a proliferative disease such as cancer or a non-oncologic disease. In some embodiments, a mutant kinase or fusion protein is associated with resistance to a kinase inhibitor (e.g. the mutant or fusion protein kinase is less effectively inhibited by a kinase inhibitor than a corresponding non-mutated or non-fused version). In some embodiments a drug resistance mutation is a mutation that arises, e.g., in a subject being treated for cancer with a kinase inhibitor, wherein the mutation renders a tumor that was initially susceptible to therapy with the kinase inhibitor no longer sensitive to such therapy. In general, a tumor may be considered sensitive to therapy if administration of the kinase inhibitor results in regression of the tumor, e.g., an objective response to therapy, or at least a stabilization or slowing of progression. A tumor may be considered resistant if it recurs in the presence of therapy or continues to progress in the presence of therapy. Numerous kinase mutations found in human cancer are known in the art, some of which contribute to cancer, and some of which render tumor cells resistant to kinase inhibitor therapy. See, e.g., Appendix I - XVI in Matthews, DJ and Gerritson, M., Targeting Protein Kinases for Cancer Therapy, Wiley, (2010), available online at ftp://ftp.wiley.com/public/sci tech med/protein kinase. In some embodiments, a mutation alters the activity of the protein. A mutation may result in a protein with increased activity (an activating mutation), decreased activity, altered localization, altered regulation, etc.

[00129] A. Viruses

[00130] Any of a variety of viruses capable of infecting eukaryotic cells, e.g., mammalian or avian cells, may be used in arrays of the present invention in various embodiments. For purposes of convenience, the process by which viruses are introduced into cells may be referred to herein as "transfection" or "infection", and the cells that have taken up viruses may be referred to as being "transfected" or "infected". However, it should be understood that the virus need not be capable of carrying out the complete infectious cycle, and it should be understood that "transfection" or "infection" does not imply any particular mechanism of viral entry. Typically infection of cells by viral particles involves binding to a cell surface receptor. The virus typically comprises a genome encapsulated in a surrounding envelope or capsid and is thus distinct from a DNA plasmid or other vector consisting or consisting essentially of nucleic acid. Entry of the virus into the cells introduces the nucleic acid into the cells, wherein the nucleic acid is expressed or has an effect on or interacts with a cellular component or function. In the case of some viruses, the virus is not internalized but instead "injects" its genomic material into the target cell. The virus typically comprises a nucleic acid of known sequence and/or source ("sequence of interest") that may, but need not be, incorporated into the viral genome. Once inside the cell, a copy of the viral genome or a portion thereof including the nucleic acid sequence of interest, may integrate into the genome of the cell and be inherited by progeny of the cell. In some embodiments the viral genome may be maintained as a replicable episome and inherited by progeny of the cell. In some embodiments the nucleic acid is transiently expressed in the cell following entry of the virus. After an appropriate period of time, the effect of the introduced nucleic acid(s) on cell phenotype is detected. Thus it will be understood that typically the effect on the cell of the introduced nucleic acid(s) of interest, rather than the effect on the cell of the virus per se, is of interest. Exemplary nucleic acids of interest include, e.g., cDNAs, NAi agents such as shRNAs, antisense sequences, and others described in the following section (entitled "Nucleic Acids").

[00131] The invention contemplates use of any virus known in the art to be useful for expressing heterologous nucleic acids in eukaryotic cells. In certain embodiments the virus is an enveloped virus. Examples include retroviruses (e.g., lentiviruses), adenoviruses, adeno- associated viruses, herpes viruses such as herpes simplex virus or Epstein-Barr virus, baculoviruses, measles viruses, hepatitis viruses, etc. The use of lentivirus is exemplified herein, but aspects of the invention encompass embodiments in which other viruses are used. Thus it should be understood that although various compositions and methods relating to certain aspects of the invention are described herein mainly in regard to lentiviruses, the invention provides embodiments in which any other virus or group of viruses, e.g., other retroviruses, or non-retrovirus enveloped viruses, are used.

[00132] Retroviruses are of particular use in certain embodiments in part because, following their entry into a cell, a DNA copy of the viral RNA genome is synthesized and integrates into the cell genome, which can allow for stable expression of a nucleic acid included in the viral genome. Lentiviruses are of particular interest, in part because of their ability to transduce non-dividing cells. Lentiviruses may include sequences derived from any of a wide variety of lentiviruses including, but not limited to, primate lentivirus group viruses such as human immunodeficiency viruses HIV-I and HIV -2 or simian immunodeficiency virus (SrV); feline lentivirus group viruses such as feline immunodeficiency virus (FIV); ovine/caprine immunodeficieny group viruses such as caprine arthritis encephalitis virus (CAEV); bovine immunodeficiency-like virus (BFV); equine lentivirus group viruses such as equine infectious anemia vims; and visna/maedi virus. It will be appreciated that each of these viruses exists in multiple variants or strains. Exemplary retroviruses include Moloney murine leukemia virus (MoMuLV or MMLV), Harvey murine sarcoma virus (HaMuSV or HSV), murine mammary tumor virus (MuMTV or MMTV), gibbon ape leukemia virus (GaLV or GALV), Rous sarcoma virus (RSV), avian sarcoma and leucosis virus, and spleen necrosis virus (SNV).

[00133] In general, a virus may contain (e.g., may be engineered to contain) one or more genetic element(s) derived from any of various different viral families or genera. For example, genetic elements such as the woodchuck hepatitis post-transcriptional regulatory element (WPRE) or the central polypurine tract (cPPT) are of use to enhance virus-mediated (e.g., retrovirus-mediated, adenovirus-mediated) transgene expression.

[00134] It will typically be desirable to select a virus with a tropism that allows it to infect a cell type of interest that will be used to produce a cell microarray. In general, one of ordinary skill in the art would select a virus that, either naturally or at least in part as a result of pseudotyping, is able to infect a cell type of interest. For example, the cell may express a protein or other biomolecule that serves as a receptor for the virus. Typically, a viral envelope or capsid protein (e.g., glycoprotein) binds to a receptor, which facilitates uptake of the virus. The receptor may or may not be known. In some embodiments a virus capable of infecting both human and rodent (e.g., mouse) cells is used.

[00135] In certain embodiments of the invention a virus is pseudotyped. In general, a pseudotyped virus is one in which at least some components of the outer shell (e.g., the envelope glycoproteins of an enveloped virus or the capsid proteins of a nonenveloped virus) originate from a virus that differs from the source of the genome and the genome replication apparatus. For example, a pseudotyped retrovirus may differ from its non-pseudotyped counterpart in that the envelope of the pseudotyped retrovirus incorporates a non-retro viral envelope protein, or an envelope protein from a different retrovirus, instead of, or in some cases in addition to, the native retroviral envelope protein. The two viruses may differ considerably (e.g., retrovirus and rhabdovirus), or they may be closely related (e.g., two different retroviruses or different serotypes of a virus). Pseudotyping makes it possible to alter the range of cell types and/or species that the virus can infect. Pseudotyping a viral vector can provide it with an expanded set of target cells or can restrict it to specific cells that are of experimental or therapeutic interest. A pseudotyped vector can have an altered stability and/or interaction with a host cell. For example, certain pseudotyped viral vectors can be produced and/or concentrated to higher titers than the corresponding viral vector with its native outer shell or envelope. The invention contemplates use of envelope proteins that confer any of these or other desired characteristics on a virus.

[00136] In some embodiments, a vesicular stomatitis virus (VSV)-G pseudotyped viral vector is used, as exemplified herein. Vesicular stomatitis virus (VSV) is a rhabodvirus able to infect a broad range of cell types, likely as a result of interactions of its envelope glycoprotein (VSV-G) with cell surface molecules present on these cells. It will be appreciated that VSV-G from any VSV serotype (e.g., New Jersey or Indiana) could be used. Furthermore, one of ordinary skill in the art will appreciate that VSV-G pseudotyped viral vectors may employ naturally occurring or engineered variants of VSV-G so long as the variant has the ability to mediate virus entry into cells.

[00137] In some embodiments, an inducible promoter system is used so that VSV-G (or other polypeptide expressed for pseudotyping) expression can be regulated, e.g., turned off, when it is not required (e.g., after infection). For example, the tetracycline-regulatable gene expression system (Gossen & Bujard, Proc. Natl. Acad. Sci. 89:5547-5551, 1992) and variants thereof (see, e.g., Allen, N, et al. (2000) Mouse Genetics and Transgenics: 259-263; Urlinger, S, et al. (2000). Proc. Natl. Acad. Sci. U.S.A. 97 (14): 7963-8; Zhou, X, et al (2006). Gene Ther. 13 (19): 1382-1390 for examples) can be employed to provide for inducible or repressible expression of VSV-G. Other inducible/repressible systems are known in the art and can be used in various embodiments of the invention. For example, expression control elements that can be regulated by small molecules such as hormone receptor ligands (e.g., steroid receptor ligands), metal-regulated systems (e.g., metallothionein promoter) can be used in certain embodiments. To this end, the VSV-G coding sequence may be cloned downstream from appropriate expression control element(s), such as a promoter controlled by tet operator sequences or a promoter comprising a hormone response element.

[00138] A wide range of other viral envelope glycoproteins or capsid proteins could be used in the pseudotyped viral vectors in various embodiments of the invention. Examples include viral envelope proteins from any of the afore-mentioned lentiviruses or retroviruses. Envelope glycoproteins from rhabdoviruses such as rabies virus or rabies related viruses such as Mokola or Ebola virus, alphaviruses such as Ross River virus, arenaviruses such as lymphocytic choriomeningitis virus, hepatitis B or C virus, or influenza virus, could be used. It will be appreciated that there are multiple naturally occurring strains or serotypes of some of these viruses, as well as engineered variants, and envelope proteins from any of these could be used in embodiments of the present invention. Further information pertaining to viruses may be found in Knipe, DM and Howley, PM (eds.) Fields' Virology, Lippincott Williams & Wilkins; Fifth edition, 2006.

[00139] A pseudotyped viral vector may be produced using standard recombinant DNA methods to replace a portion of the viral genome with a sequence that encodes the desired protein that is to be incorporated into the viral envelope or capsid. The introduced sequence may, but need not, replace all or part of the native envelope or capsid protein gene. The introduced sequence may be positioned so that it is operatively linked to expression control sequences such as a promoter already present in the viral genome or may include such expression control sequences. One of ordinary skill in the art will readily be able to prepare recombinant viruses of the various types mentioned herein.

[00140] In many embodiments, a virus comprises a nucleic acid to be introduced into eukaryotic cells, e.g., mammalian or avia cells. The nucleic acid may for example, encode a polypeptide or RNA of interest. In some embodiments, the nucleic acid or polypeptide inhibits expression or activity of a gene product of interest. Following introduction into cells, the nucleic acid (or a copy thereof) may integrate into the genome of the cell and be inherited by descendants of the cell. The effect of expression of the nucleic acid on cell phenotypc may be assessed. Methods of generating recombinant viruses containing a nucleic acid of interest, e.g,. a nucleic acid to be introduced into mammalian cells, are known in the art. One of ordinary skill in the art will be aware of many suitable packaging cell lines and plasmid constructs that can be used to prepare recombinant viruses. See, e.g., U.S. Pat. Nos.

6,013,516; or U.S. Pub. No. 20050251872 for discussion of exemplary retroviruses, retroviral elements, and systems and methods for preparing recombinant retroviral, e.g., lentiviral particles. Exemplary plasmid backbones for production of viruses or for introducing nucleic acids into cells include, e.g., those of pLKO plasmids (e.g., pLKO.l, pLKO.2) (The RNAi Consortium; reference 3 hereof), pLentilox plasmids (e.g., pLentilox 3.7), pLenti plaasmids (e.g., pLenti6), pGIPZ (Open Biosystems), or pSicoR (Ventura et al., Proc Natl Acad Sci U S A. (2004), 101(28):10380-5). In some embodiments, viral vectors useful for the delivery of cDNAs and/or shRNAs and/ or miRNAs described in Campeau E, et al., PLoS ONE (2009), 4(8): e6529. doi:10.1371/journal.pone.0006529, or in any of references 1-16 cited therein, may be used.

[00141] In some aspects, the invention provides methods that facilitate production of virus vector arrays. In some aspects, high throughput methods of producing concentrated virus solutions compatible with printing in a microarray format are provided that do not require ultracentrifugation. By avoiding the need for ultracentifugation, such methods make it possible, for example, to use robotic liquid handlers to produce virus solutions suitable for deposition on a surface from virus supernatants. In some aspects, the invention provides a method of preparing a virus-containing composition ("virus composition") that can be printed on a cell adhesive material and retains the ability to infect overlying cells with high efficiency. In some aspects, the invention provides a method of producing an array comprising depositing the virus composition on multiple spots that comprise a cell adhesive material to form multiple distinct features. In some aspects, the invention provides a dual purification and concentration technique that permits high-throughput, parallel preparation and subsequent printing of hundreds of unique high titer viruses within a relatively short time period, such as a single day. In some embodiments, the method yield at least approximately 80% recovery of functional virus particles. In some aspects, the invention provides a method comprising sequentially adding oppositely charged polyelectrolytes to a lentiviral supernatant to form a polymer complex. Without wishing to be bound by any theory, it is believed that the polymer complex entraps the lentiviruses by electrostatic interactions. The polymer-virus complex can then be pelleted, e.g., by low speed centrifugation, and mechanically resuspended in a desired volume of liquid. In some aspects, the invention encompasses modifying methods and/or materials useful for concentrating or purifying viruses as described in reference 14, e.g., to adapt them for high throughput virus production using, for example, small volumes and amounts of reagents (e.g., to allow use of microwell plates for virus concentration) and/or to adapt them for production of virus compositions suitable for printing on a surface, e.g., a surface comprising a cell adhesive material.

[0014 1 In some aspects, the invention relates to the discovery that viruses in polyelectrolyte-virus complexes, when resuspended in a printing buffer suitable for printing to form an array, retain high infectivity for mammalian cells. In addition, the resulting virus composition can be deposited on a cell adhesive material and reversibly affixed thereto, without substantially diminishing infectivity of the virus. The invention thus provides a virus composition comprising a polymer-virus complex and a printing buffer, wherein the composition is suitable for printing to form an array for infection of mammalian cells. In some aspects, the virus composition is highly concentrated, e.g., the virus is present at a titer of at least 10⁸ infectious units (IFU)/ml. In some embodiments, viral titer is between 10^s IFU/ml and 10¹⁰ IFU/ml, or between 10⁹ IFU/ml and 10^{1 1} IFU/ml. In some embodiments, viral titer is determined as described in reference 3, 6, or 15. A detailed protocol entitled "Relative viral titering with resazurin (alamarBlue), puromycin selection" is provided as Appendix A and is available at http://www.bi adinstitute.org/rnai/p iblic/resources/protocols. Other methods of viral titering can be used. If desired, two or more methods of viral titering can be compared using the same set of virus dilutions, and appropriate conversion factors determined. A result obtained using a particular method can be converted into a result that would be obtained if a different method had been used.

[00143] In some aspects, the invention provides a virus composition comprising (a) an enveloped virus; and (b) first and second oppositely charged polyelectrolytes, i.e., an anionic (negatively charged) polyelectrolyte and a cationic (positively charged) polyelectrolyte. In some embodiments, the enveloped virus is a retrovirus, e.g., a lentivirus. In general, the cationic polyelectrolyte and anionic polyelectrolyte cause the virus to precipitate from the liquid composition when it is centrifuged, e.g., at low speed. In some embodiments, the composition further comprises a printing buffer. In some embodiments, the invention provides a method for preparing a virus composition for printing on a surface, the method comprising: (a) providing a liquid composition comprising a virus; (b) combining the liquid composition with an anionic polyelectrolyte and a cationic polyelectrolyte; (c) centrifuging the resulting composition to obtain a pellet comprising the virus and a supernatant; (d) separating the supernatant from the pellet; and (e) resuspending the pellet in a printing buffer suitable for depositing the virus on a surface. In some embodiments, the pellet is directly resusponded in a printing buffer, e.g., the printing buffer is added to the pellet and the pellet is resuspended therein using, e.g., mechanical agitation. In some embodiments the pellet is first resuspended in a suitable liquid vehicle that does not adversely affect viral infectivity, and one or more substances are added to the resulting composition so as to produce a printing buffer-virus composition. In some embodiments, the one or more additional substances are added sequentially (in any order in various embodiments). In some embodiments, at least some of the additional substances are added together. [00144] The liquid composition of step (a) can be, e.g., tissue culture medium

(supernatant) harvested from a virus packaging cell line, which may be referred to as a "virus stock". The volume of the buffer used to resuspend the viruses can be selected based, e.g., on the final concentration of virus desired to provide good infectivity when printed on a cell adhesive material or in a composition comprising a cell adhesive material. In some embodiments, the pellet is resuspended in a volume that is about 10- to 100-fold less than the initial volume of virus stock, so that the final concentration of the viruses is about 10- to 100- fold greater than the concentration of the viruses in the virus stock. In some embodiments, the pellet is resuspended in a volume that is about 10- to 20-fold less than the initial volume of virus stock. In some embodiments, the volume is selected so as to result in a virus titer of at least 10⁸ IFU/ml. In some embodiments, the method further comprises: (f) depositing at least a portion of the virus-printing buffer composition formed in step (e) onto a surface. In some embodiments the composition is deposited on a cell adhesive material on said surface. In some embodiments, prior to step (f), at least a portion of the composition of step (e) is transferred to a different vessel. In many embodiments, step (a) comprises providing the liquid composition in a well of a multiwell plate (e.g., a 96 or 384 well plate), and steps (b), (c), (d), and (e) are performed without removing the virus from the well. In many embodiments, multiple virus-printing buffer compositions are thereby prepared in parallel. Multiple plates can be processed simultaneously or within a relatively short time period, e.g., a single day, thereby allowing for rapid preparation of hundreds to thousands of distinct virus agents. "Multiwell plate", also termed "microplate", "microwell plate", "microtiter plate", etc.) refers to any container that contains a plurality of wells or vessels suitable for holding solutions in discrete, defined locations, typically in a planar grid of mutually perpendicular rows and columns. The wells can be, e.g., flat-bottomed, round-bottomed, or conical- bottomed in various embodiments. The walls of the wells may be perpendicular to the bottom or may be sloping in various embodiments. In some embodiments a microwell plate conforms to the microplate standards developed by the Society for Biomolecular Sciences (SBS) (now SLAS), published by the American National Standards Institute (ANSI)

[00145] Any of a variety of different positively and negatively charged polyelectrolytes can be used in various embodiments. In general, any of a wide variety of positively and negatively charged polyelectrolytes capable of complexing with each other can be used. In many embodiments, the positively and negatively charged polyelectrolytes are polymers. In some embodiments the polyelectrolytes are positively or negatively charged when present in a liquid composition at, e.g., a pH between about 6 and 9, e.g., about 6.5 - 8.5, e.g., about 7- 8. In some embodiments a liquid composition comprises phosphate buffered saline (PBS) or another physiologically acceptable buffered solution. In some embodiments, the anionic polyelectrolyte is a glycosaminoglycan or polysaccharide. In some embodiments the glycosaminoglycans and polysaccharides are sulfated. In some embodiments the anionic polyelectrolyte is a chondroitin sulfate, heparin, heparan sulfate, keratan sulfate, poly(amino acid), or synthetic anionic polymer. In some embodiments the anionic polyelectrolyte is a poly-L-glutamic acid, poly-L-aspartic acid, poMglycolic acid), polyflactic acid), or poly(lactic-co-glycolic acid) (PLGA). In some embodiments the cationic polyelectrolyte is (diethylamino)ethyl dextran, histones, protamine, poly-L-arginine, poly-L-histidine, or poly- L-lysine. In some embodiments the cationic polyelectrolyte is polybrene (hexadimethrine bromide, also known as l ,5-Dimethyl-l ,5-diazaundecamethylene polymethobromide). In various embodiments the chondroitin sulfate is chondroitin sulfate A (chondroitin-4-sulfate), chondroitin sulfate C (chondroitin-6-sulfate), chondroitin sulfate D (chondroitin-2,6-sulfate), chondroitin sulfate E (chondroitin-4,6-sulfate), or a mixture comprising two or more of these. In some embodiments the anionic polyelectrolyte is chondroitin sulfate and the cationic polyelectrolyte is polybrene.

[00146] The polyelectrolytes can be provided in an amount sufficient so as to result in a suitable concentration in the composition. In some embodiments, the cationic and anionic polyelectrolytes are present in the composition at approximately the same concentration. In some embodiments different concentrations of the polyelectrolytes can be used. For example, concentrations may differ by, e.g., a factor of up to about 2-3-fold in various embodiments. In some embodiments polyelectrolytes are each present in the composition at a final concentration of between 50 μg/mL and about 1 mg/ml. In certain embodiments, concentrations of about 300 μίΐ L - 500 g/mL, e.g., 400 g/mL·, are used. In various embodiments, the polyelectrolytes can be added to a virus stock or other composition comprising a virus sequentially (in either order) or together (simultaneously). In some embodiments a virus stock is added to a vessel (e.g., a well) containing one polyelectrolyte and the other polyelectrolyte is then added. The composition comprising viruses and polyelectrolytes is typically incubated for a suitable period of time. The incubation may be performed at, e.g., room temperature (e.g., about 20-23°C). Lower or higher temperatures could be used. For example, a temperature of about 37' could be used in various embodiments. Compositions may be incubated for varying period of time. For example, compositions may be incubated for about 10-20 minutes, e.g., 15 minutes at, e.g., room temperature.

|0 147] In some embodiments, "low speed centrifugation" refers to centrifugation below about 10,000 g, e.g., below about 5000 g or below about 2500g. In some embodiments, low speed centrifugation is at about 800 g - 1500 g, e.g., about 1 150g. In some embodiments polymer-virus compositions are contained in wells of a multiwell plate, e.g., a 96-well or 384- well plate (wherein, for example, each well may contain a virus comprising a distinct nucleic acid to be introduced into cells). In some embodiments, the plate is centrifuged, e.g., in a centrifuge equipped for handling multiwell plates. One of ordinary skill in the art will be able to convert g values to revolutions per minute appropriate for a given centrifuge rotor. In some embodiments, a composition comprising a polymer-virus complex is centrifuged for between 10 min and 60 min, e.g., about 15-30 min, e.g., about 20 min. One of ordinary skill in the art will appreciate that the time can vary depending, e.g., on the g value. An appropriate combination of time and g value can be selected.

[00148] Following centri fugation, the supernatant is removed (e.g., by aspiration) or the virus pellet is otherwise separated from the supernatant, and the virus pellet is resuspended. In some embodiments, virus pellet is directly resuspended in a printing buffer suitable for printing the virus onto a surface (e.g., a surface comprising a cell adhesive material). In some embodiments, the virus is first resuspended in a suitable liquid vehicle that does not adversely affect viral infectivity, and one or more substances are added to the resulting composition so as to produce a composition containing virus and printing buffer. In many embodiments, resuspension is performed at least in part by pipetting up and down multiple times, e.g., using an automated pipetting system. In some embodiments an automated pipetting system capable of simultaneously pipetting the contents of all wells of a multiwell plate, e.g., a 96 or 384 well plate is used. In some embodiments, resuspension at least partly dissociates the virus from the polyelectrolytes. In some embodiments other or additional means of at least partly dissociating the virus from the polyelectrolytes can be used. For example, in some embodiments a plate may be agitated from side to side or the contents of the wells mechanically stirred. In some embodiments an enzyme capable of at least partly degrading a polyelectrolyte can be added to a composition. In some embodiments, at least a portion of a composition comprising a virus and a printing buffer is transferred into a different vessel in preparation for printing. For example, compositions can be transferred from 96 well plates to plates with a greater number of wells, e.g., 384 well plates, in order to consolidate viruses for printing. It will be understood that only a portion of a composition is typically transferred or printed. A single well of a 96 well plate can provide sufficient composition for printing multiple arrays. In some embodiments, viruses (either suspended in a printing buffer or other composition) are stored for a period of time (e.g., days, weeks, months) prior to printing. The virus composition may, for example, be stored at low temperatures, such as 4°C or below, or frozen.

[00149] In some embodiments, viruses are concentrated by flocculation with a single cationic polyelectrolyte, e.g., a cationic polymer or a combination of two or more cationic polyelectrolytes, in the absence or substantial absence of an anionic polyelectrolyte. For example, polybrene or poly-L-lysine or other cationic polyelectrolytes mentioned above can be used. Methods used can be generally similar to the polyelectrolyte complexation methods described above, but omitting the anionic polyelectrolyte. In some aspects, the invention encompasses modifying methods and/or materials useful for concentrating or purifying viruses as described in Le Doux, J , et al, Journal of Biotechnology (2006) 125(4):529— 539, 2006 or Zhang et al, Gene Ther., (2001) 8 (22): 1745-1751 to adapt them for high throughput virus production using, for example, small volumes and amounts of reagents (e.g., to allow use of microwell plates for virus concentration) and/or to adapt them for production of virus compositions suitable for printing on a surface, e.g., a surface comprising a cell adhesive material.

[00150] In the virus concentration methods described herein, e.g., the polyelectrolyte complexation method or the cationic polymer flocculation method, concentrations, time periods, and/or centrifugation parameters can be varied and may be selected so as to achieve, for example, desirable or optimal levels of virus concentration and/or infectivity.

[00151] In some embodiments, an automated liquid handling workstation, such as a JANUS workstation (PerkinElmer) or other workstation having similar capabilities, is used to perform at least some of the steps of producing a virus composition ready for printing on a surface.

|00152] In some aspects, the invention relates to the Applicants' discovery that virus binding to a surface (e.g., reversible binding to a cell adhesive material) can be improved by purifying the virus to produce a virus composition that is substantially free of serum proteins, e.g., serum proteins that may be present in typical virus harvest medium. In some aspects, the invention provides a virus composition that is substantially free of serum. In some embodiments, serum protein concentration may be estimated based on the amount of a characteristic serum protein or proteins, such as serum albumin, alpha- 1 globulins, alpha-2 globulins, beta globulins, and gamma globulins, or particular proteins falling into any of these categories, such as transferrin, haptoglobin, alpha-2 macroglobulin, etc. For example, the concentration of a characteristic serum protein in a virus composition can be less than 10%, 5%, 1%, 0.5%, 0.2%, 0.1%, or less than the concentration of such protein in a cell culture medium comprising 10% serum (e.g., 10% fetal bovine serum). The virus composition may, for example, be prepared as described above using a negatively charged and a positively charged polyelectrolyte.

[00153] The volume of virus composition printed may vary depending, e.g., on factors such as the size of the print head (or other device) used, the desired size of the features, etc. In some embodiments between 1 nl and 10 nl of solution, e.g., between 2 nl and 5 nl of composition, may be deposited on the surface.

[00154| A variety of components may be included in a composition, e.g., a virus- containing composition or printing buffer. In some embodiments a component is commonly used in biological applications and/or is biocompatible at the concentrations in which it is present when used to print an array or during use of the array (e.g., after addition of cells and culture medium). The composition may contain a buffer compound (i.e., a compound that helps regulate the pH of the composition). In some embodiments the buffering compound has a pKa of between 7.0 and 8.0 at 25'C. In some embodiments a buffer compound is HEPES or another physiologically acceptable buffer such as MOPS, or TES. The composition may incl ude a suitable amount of a stabilizing agent or preservative. The stabilizing agent may contribute to maintaining the virus in an infectious state during the process of drying, freezing or freeze drying and/or while maintained in a low temperature state cither prior to deposition on the surface or thereafter. Suitable agents and concentrations thereof are known in the art and include carbohydrates, e.g., disaccharides such as trehalose, etc. In certain embodiments of the invention an infection-enhancing compound is included in the solution. In some embodiments an infection-enhancing compound is a compound known in the art as being useful to enhance viral infection. In some embodiments an infection- enhancing compound is a positively charged compound, e.g., a polycation such as protamine sulfate. Without wishing to be bound by any theory, such compounds may facilitate interaction between the negatively charged viral envelope and negatively charged cell surface. Other suitable components that may be present in some embodiments include DEAE-dextran, dextran sulfate, polybrene, or heparan sulfate, etc. In some embodiments, the composition contains a salt, e.g., KC1, NaC.l, etc. Other salts, e.g., those containing a monovalent or divalent cation such as Mg⁺⁺, Ca⁺⁺, etc., could also be used. The concentration of the salt may range, e.g., from 100 mM to 2 M, e.g., from 500 mM to 1.5 M, or from 1.0 M to 1,4 . Any of the various components may be present in the mixture to be deposited on the surface at a concentration of from 0,0001% to 10% (w/v), but may be present in smaller or greater amounts, e.g., from 0.001% to 10% (w/v) or from 0.01% to 1% (w/v), or from 0.1 % to 1 % (w/v). In some embodiments a printing buffer containing about 0.4M HEPES, about 1.23M KC1, about 12.5 mg/mL trehalose, and about 12 ,ug/mL protamine sulfate, pH 7.0 - 7.6, e.g., 7.3 is used.

[00155] While the polyelectrolyte-based virus complexation method described above represents a convenient and efficient means of producing high titer virus preparations in high throughput, other methods of producing high titer virus compositions can be used in various embodiments. In some embodiments a membrane-based method is used. For example, membranes having a pore size sufficiently small to avoid passage of virus particles can be installed in wells of multiwel! plates that contain a virus stock (e.g., culture supernatant han'ested from a packaging cell line) or virus stock is placed into the wells. The plates can be centrifuged, e.g., at low g values such as about 2000 g or less, and viruses thereby collected on the filter. The viruses are resuspended in a smaller volume (e.g., a 10-100 fold smaller volume) than originally contained in the well, thereby concentrating them. In some embodiments, the invention encompasses use of PEG to precipitate and concentrate viruses. See, e.g., Kohno T, J Virol Methods. (2002) 106(2): 167-73, for a description of PEG-based lentivirus concentration. In some embodiments a PEG-based method is modified to allow for its use in a high throughput context, e.g., wherein multiple viruses are concentrated in parallel, e.g., in 96 well microplates. In some embodiments, an evaporation-based method is used. Virus stock, e.g.. in culture medium in multiwell plates, is exposed to a low humidity environment (sufficiently low to result in evaporation of water from the culture medium) for a sufficient time to concentrate the virus. Optionally, a gentle stream of air or a suitable gas such as nitrogen, is passed over or directed at the plate to facilitate evaporation. Heat may be used to facilitate evaporation.

|00156] In some embodiments a virus expresses or is engineered to express a ligand on its surface, wherein the ligand is useful to affinity purify the virus. The ligand may be present as part of a fusion protein comprising at least a portion of a viral envelope protein. The ligand may be, e.g., a 6xIIis tag, epitope tag, avidin, or any other entity that can be recognized by a suitable binding partner. In some embodiments a moiety that occurs naturally on a viral surface is used as a ligand for affinity-based purification. For example, an antibody that binds to a portion of a viral envelope protein (e.g., p24) can be used as an affinity reagent. The binding partner is typically immobilized on particles (e.g., beads, such as those of a chromatography resin, magnetic beads, etc.), a membrane, an inner surface of a tube, well, channel, column, or another suitable support. The binding partner can be, e.g., Ni-NTA (for a 6xHis tag), an anti-epitope antibody (for an epitope tag such as a Myc or HA tag or a viral envelope protein), biotin (for an avidin tag), etc. Viral stock is contacted with the solid support, and becomes bound to the binding partner. The culture medium and unbound components present therein flow through or are otherwise removed, leaving the virus. Vitus is subsequently dissociated from the binding partner, e.g., by eluting in a smaller volume of liquid, e.g., printing buffer or a liquid to which printing buffer components are subsequently added. In certain embodiments, a membrane-based, affinity-based, precipitation-based, evaporation-based virus purification and concentration method is performed in microwell plates, e.g., without removing virus stock from microwell plates.

100157| In some aspects, the invention provides an array, wherein features of the array comprise viruses prepared as described herein. In some embodiments features comprise a virus- polyelectrolyte complex, wherein the complex is at least partly dissociated. It should be noted that although high throughput methods for preparation of vims compositions without ultracentrifugation are of particular interest, ultraccntrifugation-based methods can be used in certain embodiments of inventive arrays and screening methods.

[00158] It is noted that inventive high throughput virus concentration methods (e.g., using polyelectrolyte complexation in microwell plates) and resulting virus compositions can be used for any of a variety of purposes. In some aspects, the invention encompasses products and processes containing or prepared using such virus compositions or methods. For example, in some embodiments an inventive high throughput vims concentration method is used to prepare lentiviruses for use in a microwell plate based screening method, e.g., as described in reference 6, in which lentiviruses are screened in individual wells of conventional 96 or 384 well multiwell plates.

[00159] B. Nucleic Acids

[00160] In some embodiments, agents are or comprise nucleic acids. It will be understood that in some embodiments nucleic acids may be delivered by viruses, e.g., as described above. In some embodiments, nucleic acids are not contained within viruses. [00161] The nucleic acids may, for example, be linear or circular DNA, RNA, or a DNA/RNA hybrid (which may in each case be double-stranded, single-stranded, or partly double-stranded) in various embodiments. A linear double-stranded nucleic acid molecule could be blunt-ended or may have an overhang at either or both ends (e.g., a 3' overhang).

[00162] The nucleic acids may be from any of a variety of sources, such as nucleic acid isolated from cells, or may be recombinantly produced or chemically synthesized. Where the source of sequences is naturally occurring, those sequences can be isolated from any cell or collection of cells in various embodiments. For instance, the sequences can be isolated from the cells of either adult tissue or organs or embryonic tissue or organs at any given developmental stage (including oocyte, blastocyte, etc.). The cells can be derived from healthy tissue or diseased tissue. In the case of a solid organ, the cell sample can be obtained by, e.g., biopsy. For blood, lymph and other bodily fluids, the cells can be isolated from the fluid component, e.g., by filtration, affinity purification, centrifugation or any other technique known in the art.

[00163] In some embodiments, the array can include coding sequences from cDNAs or genomic DNA. Where reference is made herein to cDNAs or ORFs, it should be understood that the invention encompasses embodiments in which sequences containing one or more introns are used. In some embodiments, the coding sequences can include sequences that have been mutated relative to the native sequence, e.g., a coding sequence that differs from a naturally occurring sequence by deletion, substitution or addition of at least one residue. A nucleic acid can correspond to full length or partial sequences, can be antisense in orientation, or can comprise or consist of a non-coding sequence.

[00164] In some embodiments, all or a portion of a nucleic acid sequence can be synthesized chemically. Random and semi-random sequence can thus be introduced into the nucleic acids. Modified forms of nucleotides and nucleotide linkages may be used in some embodiments.

[00165] The nucleic acids can be present as part of a larger vector, such as an expression vector (e.g., a plasmid or viral-based vector), but need not be. The nucleic acids of the array can be introduced into cells in such a manner that at least part of the introduced sequence (e.g., including the sequence of interest becomes integrated into the genomic DNA and is expressed, or in some embodiments, such that the sequence remains extrachromosomal (e.g., is maintained episomally). |00166] In general, nucleic acids for use in the arrays of the invention can be of any size. In certain embodiments, e.g., where expression vectors such as plasmids are used, a sequence of interest can be from about 30 nucleotides (nt) to about 10 kb in size, e.g., about 50 nt to about 5 kb, e.g., about 200 nt to 2 kb. In such embodiments, the arrayed nucleic acid, e.g., which includes the expression vector backbone as well as a sequence of interest, can be from about 1 kb to about 15 kb, e.g., from about 5 kb to about 8 kb.

100167] In certain embodiments, the agents on an array comprise a library of expression vectors comprising diverse sequences to be expressed. Ligating a polynucleotide coding sequence or other transcribable sequences into an expression vector can be carried out using standard procedures.

[00168] In some embodiments, it will be desirable that the vector be capable of replication in the cell. It may be a DNA which is integrated into the host genome, and thereafter is replicated as a part of the chromosomal DNA, or it may be DNA which replicates autonomously , as in the case of an episomal plasmid. In the latter case, the vector will include an origin of replication which is functional in the host. In the case of an integrating vector, the vector may include sequences that facilitate integration, e.g., sequences homologous to host sequences, or encoding integrases. The use of retroviral long terminal repeats (LT ) or adenoviral inverted terminal repeats (ITR) in the construct can, for example, facilitate the chromosomal integration of the construct.

[00169] Appropriate cloning and expression vectors for use with mammalian, avian, insect, fungal, and other eukaryotic cells are known in the art and may be used in the present invention. The expression vectors may comprise non-transcribed elements such as an origin of replication, suitable expression control sequences (e.g., a suitable promoter and/or enhancer) linked to the portion of the nucleic acid to be expressed, and other 5' or 3' flanking nontranscribed sequences, and 5' or 3' nontranslated sequences, such as ribosome binding sites, a poly-adenylation site, splice donor and acceptor sites, and transcriptional termination sequences. As known in the art, many mammalian expression vectors contain both prokaryotic sequences, to facilitate the propagation of the vector in bacteria, and one or more eukaryotic transcription units for expressing a sequence of interest in eukaryotic cells. Numerous such vectors are known in the art. Examples include, e.g., the pcDNA vector series, pSV2 vector series, pCMV vector series, pRSV vector series, pEFl vector series, Gateway ® vectors, etc. Examples of baculovirus expression systems useful for transfection of insect cells include, e.g.,pVL-derived vectors (such as pVL1392, pVL1393 and pVL941), pAcUW-derived vectors (such as pAcUWl), and pBlueBac-derived vectors (such as pBlueBac III). For other suitable expression systems for both prokaryotic and eukaryotic cells, as well as general recombinant procedures, see Sambrook, supra, and Ausubel, supra. (00170] In some embodiments, a nucleic acid encodes a selectable marker or detectable label. In many embodiments, a nucleic acid that encodes a selectable marker or detectable label also contains a sequence of interest.

[00171] In certain embodiments, sequences of interest are cDNA sequences derived from mRNA isolated from a cell or cells of interest. A variety of methods are known in the art for isolating RNA from a cellular source, any of which may be used in embodiments of the instant invention. In certain embodiments, the subject array can be made of a library of related, mutated sequences, such as a library of mutants of a particular protein, or libraries of potential promoter sequences, etc. In another embodiment, the array provides a library of small gene fragments as the sequences of interest, e.g., sequences which may encode dominant-acting synthetic genetic elements (SGEs), e.g., molecules that interfere with the function of genes from which they are derived (antagonists) or that are dominant constitutive ly active fragments (agonists) of such genes. A library comprising known SOEs can be used, or SGEs can be identified using an array of the invention. For cDNA-derived libraries, the nucleic acid library can be a normalized library containing roughly equal numbers of clones corresponding to each gene expressed in the cell type from which it was made, without regard for the level of expression of any gene.

[00172] In some embodiments, an array comprises a library of nucleic acids encoding a diverse population of small peptides, e.g., 4-25 amino acid residues in length. The library can be generated from coding sequences of total cDNA, or single genes, or can be random or semi-random in sequence. In some embodiments, the subject method is carried out with an array in which a sequence introduced into a cell is transcribed in the cell and gives rise to double stranded RNA, e.g., shRNAs, siRNA, miRNA precursors. It will be understood that complementary portions of the sequence can hybridize in the cell and may undergo further processing, e.g., by Drosha, Dicer, or other components of the RNAi machinery, to produce an siRNA or miRNA.

[00173] In some embodiments, if secretion of a polypeptide by the cells is desired, a secretion signal sequence can be included as part of a fusion protein. In some embodiments, a sequence appropriate to direct a polypeptide to a particular cell organelle (e.g., the nucleus, mitochondria, etc.) is included. Secretion may be desirable if, for example it is desired to assess the effect of a polypeptide on the phenotype of neighboring cells (e.g., cells located atop the same feature). In many embodiments, vectors contain regulatory elements that direct expression in mammalian cells, such as the cytomegalovirus (CMV) promoter, EF1 alpha promoter, ubiquitin promoters (e.g., ubiquitin B or C promoter), SV40 promoter, etc. In some embodiments, a promoter that ordinarily directs transcription by a eukaryotic RNA polymerase I (a "pol I promoter") is used. In some embodiments, a promoter that ordinarily directs transcription by a eukaryotic RNA polymerase II (a "pol II promoter") is used. In some embodiments, a promoter that ordinarily directs transcription by a eukaryotic RNA polymerase III (a "pol III promoter") is used. One of ordinary skill in the art will select an appropriate promoter for directing transcription. For example, a pol I or III promoter (e.g., an HI or U6 promoter) is often used in the art for expression of relatively short R As such as shRNAs (although pol II promoters may be used).

[00174] In some embodiments, constitutive expression control sequences (e.g., constitutive promoters) are used. A constitutive promoter is one that typically directs expression in a variety of different cell types and/or does not require the presence of particular conditions or compounds to cause it to become active. Typical constitutive promoters include, for example, those that direct expression of "housekeeping" genes in eukaryotic cells. In some embodiments, regulatable (e.g., inducible or repressible) expression control sequences are used, e.g., such as the Tet system, small molecule-inducible, or metal-inducible promoters. In some embodiments, the invention contemplates use of site-specific recombination (which term refers to the enzyme-mediated cleavage and ligation of two defined polynucleotide sequences.). In some embodiments, a nucleic acid, e.g., a vector, comprises sites for cleavage by a site-specific recombinase. Site-specific recombinase systems include, e.g., the Lox Cre, Flp/Frt systems. For example, at least a portion of a nucleic acid may be flanked by recombinase sites (e.g., LoxP sites). The use of site-specific recombination can allow, for example, deletion of a nucleic acid sequence located between the sites, which may, for example, result in a loss of function, or may activate expression (e.g., by deleting a "stuffer sequence" and thereby bringing a regulatory sequence such as a promoter into operable association with a sequence to be transcribed). In some embodiments, cells to be plated on an array are genetically engineered to inducibly express a recombinase such as the Cre recombinase (e.g., using a Tet system). When desired, after plating the cells, a small molecule (such as doxycycline) can be added to the culture medium to induce expression of the recombinase and thereby cause site-specific recombination to occur. In some embodiments a recombinase can be delivered to the cells exogenously after plating, or cells can be transfected after plating with an expression vector providing expression of the recombinase. It will be understood that variant LoxP sites exist and may be used in various embodiments. It will be understood that variants Cre enzymes exist and may be used in various embodiments. For example, Cre (or at least an active portion thereof) can be fused with a ligand-binding domain of a steroid hormone receptor (e.g., the estrogen receptor). Upon exposure to an appropriate ligand (which may be added to culture medium), the fusion protein can enter the nucleus. In some embodiments a ligand-binding domain is a mutated ligand-binding domain.

[00175] In certain embodiments, the array provides multiple different sequences of interest in at least some of the features, e.g., in order to perform co-transfection of the cells with at least two different sequences of interest. Co-transfection refers to the introduction of two or more plasmids or other DNA or nucleic acid constructs into the same cell. Co-transfections can be performed if a feature contains more than one plasmid or nucleic acid construct. In some embodiments features can include, for example, 2-10 different sequences of interest per feature. In some embodiments, multiple different sequences of interest are contained in a single vector, optionally under control of distinct promoters. In some embodiments the identity of a first sequence of interest is the same among different features while the identity of a second sequence of interest is varied. The capacity to co-transfect cells has many important uses. Such uses can include, for example, the ability to: infer the expression of a gene product by detecting the expression of a co-transfected nucleic acid encoding a marker protein; express multiple components of a multi-subunit complex in the same cell; express multiple components of a signal transduction pathway (e.g. MAP kinase pathway) or other pathway of interest, in the same cell; express multiple components of a pathway that synthesizes a small molecule or other molecule of interest in the same cell, express multiple RNAi agents targeting either the same gene or different genes; express combinations of genes that act together or in series to promote differentiation along a certain tissue lineage;

implement assays to detect protein-protein interactions; identify genetic interactions, Exemplary assays of use to detect protein-protein interactions include, e.g., mammalian two- hybrid assays in which plasmids encoding bait and prey proteins are co-transfected into the same cell, "protein fragment complementation assays" (PCA), in which a reporter molecule (typically a protein) capable of generating a detectable signal is reconstituted as a result of interaction between two or more proteins, each of which comprises a fragment of the reporter molecule, often at the N- or C- terminus. In a PCA, reconstitution of the reporter molecule results, e.g., in a protein that can be directly or indirectly detected. Fragments are selected that produce no or low signal by themselves and have low affinity for each other but have the capacity to reassemble to form a detectable reporter molecule when brought into proximity. The sequence of a fragment of a reporter molecule can be altered to, e.g., reduce spontanenous assembly of the fragments. Examples of PCAs include enzyme complementation assays, fluorescence complementation assays, luciferase complementation assays, and protease complementation assays. Exemplary reporter proteins of use in PCAs include enzymes such as dihydrofolate reductase and β-lactamase; fluorescent proteins such as green fluorescent protein (GFP) and variants thereof; and luciferases such as firefly luciferase, Gaussia luciferase, and Renilla luciferase. The split tobacco etch virus (TEV) protease assay is an exemplary protease complementation assay.

[00176] In some embodiments, a nucleic acid, e.g., a DNA, is deposited in a composition comprising a cell adhesive material, such as gelatin, thereby producing cell adhesive features comprising the nucleic acid. The composition typically comprises water and may further comprise, e.g., a sugar such as sucrose, a buffer solution that facilitates DNA molecule condensation (e.g., as described in U.S. Pat. No. 6,544,790 or 6,951,757). A suitable transfection reagent, e.g., a lipid-based transfection reagent, can subsequently be applied, e.g., prior to or after seeding the array with cells. In some embodiments a transfection reagent is applied prior to seeding the array with cells. The array is maintained for sufficient time to allow complex formation between the transfection reagent and the DNA. After a sufficient time (e.g., about 20 minutes at 25°C) for complex formation to occur, transfection reagent is removed, producing a surface bearing DNA. Cells in an appropriate medium are added. The resulting product (a surface comprising plated cells and features that comprise DNA) is maintained under conditions that result in entry of DN A into plated cells, thus producing an array (a surface bearing an array) of reverse transfected cells that contain defined DNA and are in discrete, defined locations on the array. Optionally, the array is allowed to dry prior to seeding with cells.

[00177] Γ^η some embodiments, a cell adhesive material is first deposited to form cell adhesive regions, followed by deposition of the nucleic acids onto said regions. In some embodiments, the nucleic acids are deposited on ceil adhesive regions in a composition comprising one or more lipids and the nucleic acid, wherein the lipid(s) facilitate transfection of the nucleic acid into eukaryotic cells subsequently seeded onto the array. In some embodiments, the nucleic acid is deposited on cell adhesive regions in a suitable printing buffer substantially lacking lipids. A transfection reagent is subsequently added, as described above.

[00178] Numerous transfection reagents are known in the art and may be used in embodiments of the invention. Examples of commercially available transfection reagents include, e.g., Lipofectamine, Effectene, Polyfect, and numerous others. In some embodiments a non-lipid based transfection reagent may be used. For example, polyamine-based transfection reagents are available. See U.S. Pat. No. 6,544,790 or 6,951 ,757 for further discussion of certain reverse transfection arrays and exemplary nucleic acids.

[00179) C. Small molecules

[00180] In some embodiments, an agent is a small molecule. Any suitable method can be used to produce a small molecule array wherein the array comprises means effective to substantially confine subsequently deposited mammalian cells to features comprising the small molecules. In certain embodiments small molecules are affixed to a surface by means of a biocompatible matrix that immobilizes the small molecule to the surface and prevents immediate release of the molecule from the surface, permits release of the small molecule at an appropriate rate under the conditions under which an assay is carried out. The biocompatible matrix is typicaliy deposited on the surface to form features of the array using, e.g., a multi-pin or multi-jet printhead. In some embodiments the biocompatible matrix may be a semi-permeable or biodegradable polymer. In some aspects, the invention contemplates use of products (e.g., materials) and/or methods described in I'd application

PCT/US2002/021 72 (WO/2003/056293) SMALL MOLECULE MICRO ARRAYS. Any of a wide variety of polymers useful for drug delivery (e.g., sustained or controlled drug delivery systems) can be used in various embodiments. For example, PLGA can be used. In some embodiments, the biocompatible matrix comprises a cell adhesive material. For example, gelatin, collagen, or other cell adhesive material may be used. In some embodiments the biocompatible matrix comprises a mefnacrylate-based polymer, a polycarboxylic acid, a cellulosic polymer, polyvinylpyrrolidone, maleic anhydride polymer, polyamide, polyvinyl alcohol and polyethylene oxide, A matrix can be in the form of an individual article or may comprise multiple microparticles or nanoparticlcs. In some embodiments a disk-shaped matrix is used. In some embodiments a cell adhesive material is deposited atop a matrix comprising a small molecule. In some embodiments a matrix comprising a small molecule is deposited on top of cell adhesive material. In some embodiments, a feature i s formed by depositing small volumes of the matrix at multiple locations on a cell adhesive region so that portions of the cell adhesive region are not covered by the matrix and remain exposed for adherence of subsequently deposited cells. In some embodiments, a small molecule is deposited within a layer of cell adhesion resistant material. The small molecule may be deposited in a depression in said layer. The subsequent addition of a cell adhesive material traps the molecule in the depression. The molecule subsequently diffuses through the cell adhesive material to reach overlying cells. In some embodiments, a small molecule, optionally in a biocompatible matrix, is deposited into depressions in the substrate surface (e.g., depressions in the surface of a glass slide) or in microwells in a slab of a flexible, moldable material such as PDMS. In some embodiments the slide or slab will already (i.e., prior to depositing the small molecules) comprise or have had a cell adhesion resistant material applied thereto in regions outside the wells. Λ cell adhesive material is applied atop the wells. The array is subsequently seeded with cells. The small molecules diffuse through the cell adhesion resistant layer and the cell adhesive material to reach the cells. In other embodiments a cell adhesion resistant material is applied to the substrate or slab, followed by printing with a cell adhesive material. The array is subsequently seeded with cells. The small molecules diffuse through the cell adhesion resistant layer and the cell adhesive material to reach the cells. In some embodiments cell adhesive regions are formed by depositing a composition comprising a cell adhesive material and a small molecule.

[00181] In some embodiments a small molecule is present at multiple different concentrations. In some embodiments the invention provides arrays comprising features that contain both a small molecule and a second agent (e.g., a nucleic acid that confers a loss of function of one or more genes or results in expression of a polypeptide).

[00182] In general, any small molecule can be used in embodiments of the invention relating to small molecules. Small molecules may be naturally occurring or invented by man. They may be at least in part chemically synthesized or, if naturally occurring, purified from natural sources.

[00183] In some embodiments, a compound collection ("library") is used. The library may comprise, e.g., between 100 and 500,000 small molecules, or more. Compound libraries are often arrayed in multwell plates. They can be dissolved in a solvent (e.g., DMSO) or provided in dry form, e.g., as a powder or solid. Compound libraries can comprise structurally related, structurally diverse, or structurally unrelated compounds. In some embodiments, a library comprises at least some compounds that have been identified as "hits" or "leads" in a drug discovery program and/or analogs thereof. A library can be focused (e.g., composed primarily of compounds having the same core structure or scaffold, derived from the same precursor, or having at least one biochemical activity in common. In some embodiments, compounds that have been identified as inhibitors of one or more proteins of interest, e.g., one or more enzymes are used. For example, a collection of kinase inhibitors could be used.

[00184] Compound libraries are available from a number of commercial vendors such as Tocris Bioscience, Nanosyn, BioFocus, and from government entities. For example, the Molecular Libraries Small Molecule Repository (MLSMR), a component of the U.S. National Institutes of Health (NIH) Molecular Libraries Program is designed to identify, acquire, maintain, and distribute a collection of >300,000 chemically diverse compounds with known and unknown biological activities for use, e.g., in high-throughput screening (HTS) assays (see https://mli.mh.gov/mh7). The NIH Clinical Collection (NCC) is a plated array of approximately 450 small molecules that have a history of use in human clinical trials. These compounds are highly drug-like with known safety profiles. The NCC collection is arrayed in six 96-well plates. 50 μΐ of each compound is supplied, as an approximately 10 mM solution in 100% DMSO.

[ 00185] In some embodiments, a collection of compounds comprising "approved human drugs" is tested. An "approved human drug" is a compound that has been approved for use in treating humans by a government regulatory agency such as the US Food and Drug Administration, European Medicines Evaluation Agency, or a similar agency responsible for evaluating at least the safety of therapeutic agents prior to allowing them to be marketed, A compound may be, e.g., an antineoplastic, antibacterial, antiviral, antifungal, antiprotozoal, antiparasitic, antidepressant, antipsychotic, anesthetic, antianginal, antihypertensive, antiarrhythmic, antiinflammatory, analgesic, antithrombotic, antiemetic, immunomodulator, antidiabetic, lipid- or cholesterol-lowering (e.g., statin), anticonvulsant, anticoagulant, antianxiety, hypnotic (sleep-inducing), hormonal, or anti-hormonal drug, etc. See, e.g., Goodman and Oilman's The Pharmacological Basis of Therapeutics, Twelfth Edition, McGraw Hill, 2010, for examples of therapeutically useful compounds. In some embodiments, a compound is one that has undergone at least some preclinical or clinical development or has been determined or predicted to have "drug-like" properties. For example, the test compound may have completed a Phase 1 trial or at least a preclinical study in non-human animals and shown evidence of safety and tolerability. In some embodiments, a test compound is substantially non-toxic to cells at the concentration to which cells would be exposed in a screen using an array of the invention. For example, there may be no statistically significant effect on cell viability and/or proliferation, or the reduction in viability or proliferation can be no more than 1%, 5%, or 10% in various embodiments. Cytotoxicity and/or effect on cell proliferation can be assessed using any of a variety of assays (some of which are mentioned elsewhere herein). In some embodiments, cytotoxicity and/or effect on cell proliferation is tested using an array of the invention. In some embodiments at least 80% of the small molecule on an array fulfill at least one of the foregoing criteria.

[00186] In some embodiments, one or more compounds or mixtures thereof having a known activity (e.g., a therapeutic activity, an undesired activity, a cytotoxic or cytostatic or proliferation inhibitory activity) is tested, wherein the molecular target of the compound or mixture and/or mechanism of activity is unknown. Testing of such compounds or mixtures according to the present invention to determine may lead to identification of the molecular target. Such identification may, for example, facilitate development of more highly active structural analogs of the compound, and/or identification of additional compounds that act on the same target.

[00187] IV. Cells and Cell Seeding

[00188] Any of a wide variety of eukaroytic cells can be used in various embodiments of the present invention. In some embodiments, animal cells are used. In some embodiments vertebrate cells, e.g., mammalian cells, are used. For example, in some embodiments, primate cells are used. In some embodiments primate cells are human cells. In some embodiments primate cells are non-human primate cells such as monkey cells. Other mammalian cells that may be used include rodent (e.g., mouse, rat, rabbit), canine, feline, bovine, or porcine cells. In some embodiments, avian cells are used. In some embodiments insect cells, e.g., Drosophila cells are used. In other embodiments, yeast or other fungal cells are used. Cell-seeded arrays are an aspect of the present invention. In some embodiments a cell-seeded array has been maintained in culture for between 2-5 days. In some embodiments a cell-seeded array has been maintained in culture for between 5-7 days or between 7-10 days. In some aspects, a cell-seeded array has been contacted with a test compound (e.g., a test compound has been added to culture medium in which die array is maintained) during at least part of the time that the array has been maintained in culture.

[00189] Cells can be fully differentiated cells or progenitor/stem cells. They may be dividing or non-dividing cells in various embodiments. They may be primary cells or cells of non-immortalized or immortalized cell lines. Primary cell can be cells that have been obtained from a subject and have been maintained in culture for no more than 3 passages or population doublings. In some embodiments, cells have been maintained in culture for at least 5 population doublings or passages, or at least 10 population doublings or passages. In some embodiments, a "cell line" refers to a population of cells that has been maintained in culture for at least 10 passages or 10 population doublings. It will be understood that the cells may have been frozen down and thawed between passages. A cell line may be clonal (derived from a single cell) or polyclonal (derived from multiple cells). In many embodiments a polyclonal cell line will have been derived from a sample of cells obtained from a particular individual and will contain cells that are genetically identical or close to identical. An immortalized cell line is a cell line that has acquired an essentially infinite life span, i.e., the cell line is capable of proliferating indefinitely. For purposes of the present invention a cell line that has undergone or is capable of undergoing at least 100 population doublings in culture is considered immortal. A non-immortalized cell line may, for example, be capable of undergoing between about 20-80 population doublings in culture before senescence. In some embodiments, an animal cell line is a transformed cell line. As used herein, an animal cell line is considered transformed if it exhibits: (a) immortalization, (b) aberrant growth control, evidenced by loss of at least one of the following: contact inhibition of cell motility, density limitation of cell proliferation, and anchorage dependence, and, in some embodiments, (c) malignancy, as evidenced by the growth of invasive tumors in vivo (e.g., in appropriate isogenic or immunocompromised host animals, e.g.,

immunocompromised mice).

[00190] Numerous eukaryotic cell lines are known in the art. Many such cell lines can be obtained from cell banks and repositories such as the American Type Culture Collection (ATCC), Deutsche Sammlung von Mikroorganismen und Zellkulturen (German Collection of Microorganisms and Cell Cultures; DSMZ), European Collection of Cell Cultures (ECACC), Japanese Collection of Research Bioresources (JCRB), RIKEN, Cell Bank Australia, etc., the catalogs (e.g., paper or online catalogs) of which are incorporated herein by reference. In some embodiments, a mammalian cell line is A375, Colo679, UACC-62, Malme-3M, WM- 793, WM-1716, WM-1745, WM-1852, WM-1930, Sk-Br-3, HCC-827, UACC-812, ZR-75-1 , WM-3627, WM-451Lu, WM-1862, WM-3163, Sk-Mel-28, Sk-Mel-5, Lox IMVI, IGR-39, Hs294T, A2058, A549, U2-OS, U87, SW620, HBL-100, MCF-7, PC3, MDA-MB-231, MDA-MB-453, SW480, HCT-1 16, DDLS8817, LPS141, HeLa, 786-0, HepG2, DU145, 90- 8T, 293T, Panc-1, WM-115 or RPMI-7951, See Table 12,1 of Freshney (2010) for additional exemplary mammalian cell lines that may be used in certain embodiments. In some embodiments a mammalian cell line listed in Supplementary Table 2 is used.

[00191 J In some embodiments, cells are epithelial cells. In some embodiments, cells are mesenchymal cells. In some embodiments, cells are adult stem cells or embryonic stem cells. In some embodiments, cells are induced pluripotent stem (iPS) cells, e.g., human or rodent iPS cells. Induced pluripotent stem cells are somatic cells that have been "reprogrammed" to a pluripotent state. Such cells may be generated using any of a variety of methods. In general, such methods often involve expressing, causing expression of, or introducing into a somatic cell at least one "reprogramming factor". Reprogramming factors include a variety of transcription factors such as Oct4, Nanog, Sox2, c-Myc, Lin28, and KLF4. For example, reprogramming may be achieved using Oct4, Sox2, Klf4, and c-Myc or Oct4, Nanog, Sox2, and Lin28 (see, e.g., Meissner, A., et al, Nat Biotechnol, 25(10): 1 177-81 (2007); Yu, J., et al, Science, 318(5858):1917-20 (2007); and Nakagawa, M., et al., Nat Biotechnol., 26(1): 101 -6 (2008). One of ordinary skill in the art will be aware of various approaches to generating iPS cells, e.g., using retroviral expression, transient transfection, small molecules, protein transduction, etc. In some embodiments, cells are derived from stem cells (which may be iPS cells), e.g., by in vitro differentiation, which differentiation may at least in part occur following seeding of the cells on an array of the invention. Cells are often dispersed in culture prior to being used to seed an array. However, in some embodiments cells may be in tissue samples that retain at least some of the microarchitecture of the tissue or organ from which they were obtained.

[00192] Cells can be derived from normal tissue or diseased tissue in various embodiments. In some embodiments diseased tissue is neoplastic tissue. In some embodiments cells are derived from tissue obtained from an individual suffering from a disease. The tissue may or may not be from a tissue showing evidence of or affected by the disease. The disease may be, e.g, a neoplastic disease, a neurodegenerative disease, a metabolic disease, a cardiovascular disease, a psychiatric disease, etc. Tissue may be embryonic, fetal, or adult tissue in various embodiments.

[00193] Exemplary cells of interest include, e.g., fibroblasts, neuronal cells, glial cells (e.g., astrocytes), pancreatic cells, hepatocytes, chondrocytes, osteocytes, osteoblasts, myocytes, myoblasts, keratinocytes, alveolar epithelial cells, bronchial epithelieal cells, cervical epithelial cells, corneal epithelial cells, esophageal epithelial cells, endothelial cells, enterocytes (columnar epithelial cells found in the small intestines and colon), mammary epithelial cells, ovarian epithelial cells, prostate epithelial cells, retinal pigment epithelial cells, melanocytes, and hematopoietic cells. Hematopoietic cells include, e.g., cells of the myeloid, erythroid, and lymphoid lineages. In some embodiments, cells are lymphocytes (e.g., T cells, B cells, NK cells). In some embodiments, cells are monocytes or macrophages. 100194] If desired, cell type may be assessed using any of a variety of approaches known in the art. For example, cell type may be assessed using gene expression profiling (e.g., using oligonucleotide or cDNA microarrays, RNA-Seq, or any other suitable method for assessing RNA expression), measuring expression of lineage or cell type specific markers (e.g., cell surface antigens, intermediate filament proteins, enzymes, hormones expressed selectively or exclusively by cells of a particular cell lineage or cell type), cell morphology, etc. A combination of characteristics may be used to identify a cell line as being of a particular type.

[001 5] In general, in order to use an array of the invention cells are plated (placed) onto a surface bearing the array in sufficient density and under appropriate conditions for interaction of the agents with the cells, e.g., for introduction/entry of a virus or nucleic acid or small molecule into the cells or interaction with a cell surface molecule (e.g., a cell surface receptor). In some embodiments, the cells (typically in an appropriate medium) are plated at a density (expressed in terms of number of cells per unit of surface area on which cells can potentially settle) of between about 0.01 - 1 x 10⁵/cm². In some embodiments the density can be, e.g., about 0.05 - 0.5 x 10⁵ /cm². In some embodiments, the density can be from about 0.05 - 0.1 x 10⁵ /cm². In some embodiments an array is seeded with about 1-5 x 10⁵ cells per area of a standard microscope slide (25 mm ^χ 75 mm). In some embodiments, cells are plated at a density (expressed in terms of number of cells/volume of culture medium) of about 0.1 -5 x 10⁵ cells/ml, e.g., about 0.2-1 x 10⁵ cells/ml. In general, the appropriate number and density of cells to be used can vary depending on factors such as the expected duration of a screen, the nature of the agents, the expected cell cycle time of the cells, etc. For example, it may be desirable to plate the cells at a concentration sufficiently low such that cells on a feature could divide up to 2-5 times without becoming confluent. Cells can be maintained on an array for a suitable period of time to perform a screen. In many embodiments cells are maintained for between 24 hours and about 10 days, e.g., 2 - 5 days, 5-7 days, 7-10 days. At least part of the culture medium may be replaced one or more times during the time period, or fresh medium added without removing medium in various embodiments. In some embodiments an agent comprises a sequence that encodes a selectable marker, and culture medium containing a selection agent (e.g., an antibiotic) is used during at least part of the time period, or other selective conditions are imposed, in order to select for cells that have taken up an agent that causes the cells to express the selectable marker. For example, cells can be allowed to adhere overnight (e.g., about 12-18 hours) and become infected or transfected by an agent, after which an antibiotic such as puromycin or blasticidin is added to the culture medium at an appropriate concentration. Without wishing to be bound by any theory, the use of a selectable marker in an agent of use in the invention may provide various advantages. In some aspects, the use of a selectable marker may eliminate or substantially reduce survival or proliferation of cells that settle on regions of the surface located between the features and might otherwise survive or proliferate even in the presence of a cell adhesion resistant material or other means to confine cells to features. In some aspects, the use of a selectable marker may eliminate or substantially reduce survival or proliferation of cells that adhere to the features but fail to be infected or transfected. The presence of such cells might, in some instances, interfere with detection of a cellular phenotype. For example, if a phenotype that results from an agent includes a reduction in proliferation or survival, the non-transfected cells (which would continue to proliferate) might make it more difficult to detect the reduced survival or proliferation of the transfected cells. In some embodiments, a better signal to noise ratio may be obtained if nontransfected cells are substantially eliminated. In some embodiments an agent comprises a sequence that encodes a detectable label (e.g., a fluorescent protein). In some embodiments an agent comprises a first sequence that encodes a selectable marker and a second sequence that encodes a detectable label. The coding sequences are under control of transcriptional control elements appropriate to direct expression in eukaryotic cells, e.g., mammalian cells.

[00196] In some embodiments, the cells are genetically engineered, e.g., prior to being used to seed an array. For example, cells can be genetically engineered to express one or more RNAs or polypeptides. In many embodiments, genetically engineered cells have a stable genetic modification (wherein the genetic modification is inherited with high efficiency by progeny of the cell and their descendants). For example, the genetic modification can comprise an insertion of heterologous DNA into the genome or a deletion of endogenous genomic DNA. A deletion of endogenous genomic DNA may, for example, functionally inactivate a gene. Cells can be genetically engineered to express any of a wide variety of RNAs. In some embodiments, cells are genetically engineered to express an RNAi agent (e.g., a microRNA or short hairpin RNA). In some embodiments cells are genetically engineered to express an mRNA that encodes a polypeptide of interest.

[00197] In some embodiments, cells are engineered with a nucleic acid that comprises a reporter gene. In some embodiments the ability of agents (e.g., nucleic acids in viral vectors) to alter the level of expression of the reporter gene can be assessed. For example, an array can be assessed to identify nucleic acids that encode or repress transcriptional activators or transcriptional repressors of the reporter gene. A "reporter" or "reporter molecule" can be any molecule whose presence or activity can be detected and, typically, quantified, and wherein the presence or activity of the molecule provides information about or serves as indicator of an event or condition of interest. Often, detection of the reporter molecule rather than directly detecting the event or condition of interest is easier, more convenient, more accurate, or has other advantages. Often a reporter molecule is a detectable protein, such as a fluorescent protein or an enzyme capable of acting on a substrate to produce a colorimetric, luminescent, or otherwise optically detectable signal. Often the event of interest is modulation (e.g., activation or inhibition) of a transcription factor or modulation of a signaling pathway that affects activity of a transcription factor. A "reporter gene" or "reporter gene construct" is a nucleic acid that comprises a portion that encodes a reporter molecule, operatively linked to at least one transcriptional regulatory sequence.

Transcription of the sequence that encodes the reporter molecule is controlled by those sequences to which it is linked. In many embodiments the activity of at least one or more of these control sequences is directly or indirectly regulated by a receptor protein, e.g., a cell surface receptor. Exemplary transcriptional control sequences are promoter sequences. Synthesis of the reporter molecule thus "reports" on the activity of the transcriptional control sequences, which may in turn reflect the activity of a signal transduction pathway. Reporter gene constructs and reporter molecules have a wide variety of uses in the context of the present invention. For example, reporter genes can be used to identify compounds that activate or inhibit a signal transduction pathway or to identify genes whose expression or inhibition modulates activity of a signal transduction pathway.

[001 8] In some embodiments, cells are engineered to have the potential to express a reporter molecule that reports on the level of activation of a signal transduction pathway. For example, the cells can comprise a reporter gene construct comprising a nucleic acid that encodes a reporter molecule, wherein the nucleic acid is operably linked to a promoter whose activation (or repression) occurs as a result of activation of signal transduction pathway. "Signal transduction" (also termed "signaling") is the processing of physical or chemical signals from the cellular environment into the cell through (across) the cell membrane, and may occur through one or more of several mechanisms, such as activation/inactivation of enzymes (such as proteases, or other enzymes which may alter phosphorylation patterns or other post-translational modifications), activation of ion channels or intracellular ion (e.g., calcium ion) stores, effector enzyme activation via guanine nucleotide binding protein intermediates, formation of inositol phosphate, activation or inactivation of adenylate cyclase, direct activation (or inhibition) of a transcriptional factor and/or activation, translocation of a protein (e.g., a transcription factor or other protein that directly or indirectly associated with DNA) into the nucleus, etc. Signal transduction pathways include, e.g., hormone signaling pathways, growth factor signaling pathways, and chemokine signaling pathways. Exemplary signal transduction pathways include, e.g., the MAPK signaling pathway, ErbB signaling pathway, Wnt signaling pathway, Notch signaling pathway, Hedgehog signaling pathway, TGF-beta-SMAD signaling pathway, mTOR signaling pathway, VEGF signaling pathway, Jak-STAT signaling pathway, NOD-like receptor signaling pathway, Toll-like receptor signaling pathway, T cell receptor signaling pathway, B cell receptor signaling pathway, calcium signaling pathway, phosphatidylinositol signaling pathway, and sub-pathways of any of the foregoing pathways.

[00199] In some embodiments, an animal cell, e.g., a mammalian or avian cell, is genetically engineered to express one or more polypeptides that immortalizes the cell. For example, the cell may be engineered to express telomerase reverse transcriptase (also termed the "telomerase catalytic subunit", e.g., human telomerase reverse transcriptase (hTERT) or, in some embodiments, various viral genes such as SV40 large T antigen, adenovirus El a, human papilloma virus (HPV) E6 and E7, and Epstein-Barr virus (EBV; e.g., the whole virus may be used).

[00200] In some embodiments, a cell is engineered to be tumorigenic. In some embodiments, a cell is engineered to express an oncogene or to lack expression of a tumor suppressor gene,

|00201] In some embodiments, a cell is engineered to express a protein that is a target of a compound. In some embodiments, the protein is a variant protein that is at least partly resistant to such compound, e.g., the compound has a reduced effect on the variant protein as compared with the effect of such compound on the protein. [00202] In some embodiments, cells can be engineered so as to have a loss of function or gain of function phenotype. The ability of one or more members of a library of agents to modify, e.g., to counteract, such a phenotype may be assessed.

[00203] In some embodiments, cells express a cell surface receptor of interest. In some embodiments, cells are engineered to express a recombinant cell surface receptor. The ability of one or more members of a library of agents to induce or inhibit signal transduction by the receptor may be assessed. "Cell surface receptor" encompasses cellular molecules (typically proteins) that are at least partially exposed on the surface of cells, interact with the extracellular environment, and transmit or transduce information regarding the environment intracellularly in a manner that may, for example, modulate intracellular second messenger activities or transcription regulated by specific expression control elements (e.g., promoters), often resulting in transcription of specific genes. Binding of a ligand to a cell surface receptor may activate a signal transduction pathway leading to such modulation or transcription.

[00204] A population of cells used to seed an array can be phenotypically uniform (e.g., may consist of cells that are all of the same cell line or cell type) or can comprise cells having detectably distinct phenotypes (e.g., cells of different cell lines or different cell types). For example, a mixture composed of cells of two more different cell lines can be used. If a mixture of different cell lines or cell types is used, the ratios of the different cell lines or cell types can vary. In some embodiments, a cell population comprises at least 50%, 60%, 70%, 80%, 90%, 95%, 96%, 97%, 98%, 99%, or more cells of a first cell line or cell type, with the remaining cells being of a second cell line or cell type, or being of multiple different cell types. In some embodiments, a population of cells comprises two or more subpopulations, wherein cells of the subpopulations are substantially identical but differ with regard to one or more selected phenotypic or genotypic characteristics.

[00205] In some embodiments, cells are labeled prior to or after being plated onto an array. In some embodiments, cells are metabolically labeled. In some embodiments cells are labeled by contacting them with a substance that has affinity for one or more cellular molecules. Any of a variety of small organic molecules commonly referred to as "'dyes" or "stains" can be used to label cells. Examples of such molecules include various xanthenes (e.g., fluorescein), cyanines, naphthalenes, coumarins, oxadiazoles, pyrenes, oxazines, acridines, and derivatives of any of these, and others known to those skilled in the art. Inorganic materials suitable for use as cell labels are also available. In some embodiments, nanoparticles such as semiconductor or metal-based nanoparticles are used. Examples include quantum dots, gold particles, fluorescent particles comprising lanthanides such as europium (Eu) and terbium (Tb). See, e.g., The Molecular Probes® Handbook, cited above. In various embodiments the substance may bind to a class of biomolecule such as DNA, RNA, or protein, or may bind specifically to a particular molecule or portion thereof (e.g., a protein having a particular sequence, or a particular protein modification), set of molecules, or supramolecular structure such as a protein complex or subcellular organelle. In some embodiments, the substance may accumulate in the cytoplasm or in one or more subcellular compartment(s), Such accumulation may occur, e.g., due to diffusion, active transport, or endocytosis.

[00206] V. Detection of Cell Phenotype

[00207] Any of variety of methods (assays) can be used to detect the consequence of uptake of an agent by cells or to detect other interaction of the agent with cells in various embodiments. In a general sense, an assay provides the means for determining if the agent or a portion thereof (e.g., a nucleic acid sequence of interest) is able to confer a change in a phenotype of the cell relative, e.g., to the phenotype that would exist in the absence of the agent (e.g., the phenotype of an identical cell that lacks the introduced sequence of interest). It is noted that any assay could be considered or applied to detect or measure a "phenotype" or a "change in phenotype", and these terms are generally used interchangeably herein. Detecting a phenotype or change in phenotype may, but need not, involve a comparison between two more more conditions or states. In various embodiments the invention contemplates use of any assay known or used in the art for assessing cell phenotype and encompasses performing any such assay on cells located on an array of the invention. This can include, for example, contacting a cell array of the invention with appropriate reagent(s) for performing such assays, and detecting a cell phenotype (or change in cell phenotype). In some embodiments phenotypic changes can be detected on a gross cellular level, such as by changes in cell morphology (membrane ruffling, rate of mitosis, rate of cell death, mechanism of cell death, dye uptake, and the like). In some embodiments, the changes to the cell's phenotype, if any, are detected by more focused or specific means, such as by the detection of the level of a particular protein (such as a selectable or detectable marker), existence or amount of a protein modification (e.g., phosphorylation), or level of mRNA or second messenger, to name but a few. In some embodiments, detecting a cell phenotype (or change in cell phenotype) comprises detecting a signal (e.g., produced by a label) indicative of cell phenotype (or change in cell phenotype). Changes in the cell's phenotype can be determined by assaying reporter molecules, assaying cellular enzymes, using immunoassays, staining with dyes (e.g. DAPI, calcofluor, organelle-specific dyes), assaying electrical changes, characterizing changes in cell shape, examining changes in protein localization (e.g., nuclear translocation), protein conformation, or by counting or otherwise detecting cell number. Various phenotypic changes of interest could be detected by methods such as chemical assays, light microscopy, scanning electron microscopy, transmission electron microscopy, atomic force microscopy, confocal microscopy, image reconstruction microscopy, scanners, autoradiography, light scattering, light absorbance, NMR, PET, patch clamping, calorimetry, mass spectrometry, surface plasmon resonance, time resolved fluorescence, autoradiography, scintillation counting, etc. Data could be collected at single or multiple time points and analyzed by the appropriate software.

[00208] In certain embodiments, a phenotypic change comprises a difference in the growth or survival rate between cells that and those that do not. In certain embodiment a phenotype change comprises a difference in a detectable characteristic of the cells, which could include altered expression level or activity of a reporter molecule, altered staining properties, etc.

[00209] In some embodiments, immunofluorescence or other antibody-based detection methods can be used to detect a protein or post-translational modification thereof (e.g., phosphorylation). Alternatively or additionally, expression of proteins that alter the phosphorylation state or subcellular localization of another protein, proteins that bind to other proteins or to nucleic acids, or proteins with enzymatic activity can be detected. An array expressing proteins of interest could be tested for phosphorylation, sulfation, ubiquitination, glycosylation, acteylation, methylation, or other post-translational modifications by incubation with the appropriate labeling or detection reagent such as radiolabeled precursors, anti-phosphoamino acid antibodies, anti-ubiquitin antibodies, histone modification antibodies, , lectins or other specific detection reagents.

[00210] In some embodiments cells on an array are analyzed for cell phenotype while alive. In some embodiments cells are fixed, dried, or otherwise processed and rendered nonviable prior to at least some phenotypic analysis. In some embodiments cells remain substantially intact. In some embodiments the cell membrane is permeabilized, which may, for example, facilitate entry of detection reagents. In some embodiments cells are at least partly lysed, whereby the plasma membrane is substantially disrupted and intracellular contents are released in the vicinity of the cell. [00211] In some embodiments at least some cells are removed from an array prior to or following phenotype detection. Cells can be isolated from particular features or from a substantial proportion of the features. Isolated cells may be further analyzed or propagated. In some embodiments cells or lysates (or components thereof) from multiple features are transferred to a new substrate or vessel. The new substrate may be suitable for cell maintenance or may be useful to analyze the cells or cell lysate. For example, cells or lysate components may be transferred to an affinity membrane, filter, cell adhesive surface, microwell plate, etc., and may be subjected to further analysis.

[00212] When screening for bioactivity of test compounds, for example, intracellular second messenger generation can be measured. Such embodiments are useful where, for example, the arrayed library is being screened for sequences of interest that activate or inactivate or otherwise modulate a particular signaling pathway, which pathway may be affected by a test compound. A variety of intracellular effectors have been identified as being receptor- or ion channel-regulated, including adenylyl cyclase, cyclic GMP,

phosphodiesterases, phosphoinositidases, phosphoinositol kinases, and phospholipases, as well as a variety of ions.

[00213] In some embodiments, activity of a reporter molecule is assessed, e.g., to provide an indication of the activity (or modulation) of a transcription factor or signaling pathway. In some aspects, the invention provides the recognition that many phenotypes of interest, such as cell number, phosphoprotein levels, or other phenotypes that can be detected based on fluorescence, can be accurately quantified on array features by measuring only the total fluorescence intensity on each feature using a conventional DNA microarray scanner. This approach may be used, for example, as an alternative to microscope-based imaging. Thus in some embodiments, a method of the invention includes a step of gathering data relating to cell phenotype by scanning a cell array with a DNA microarray scanner. In some embodiments imaging is used, which may include use of automated image analysis software.

[00214] VI. Kits and Services

[00215] In some aspects, the invention provides kits. A kit can contain any of arrays of the present invention. The kit may include any of a number of components in addition to one or more arrays. Such components may include, e.g., any one or more of the following: cells, media, detection reagents (e.g., for detecting expression of a nucleic acid introduced into cells using the array or for detecting loss of expression of a nucleic acid or for detecting expression of a reporter molecule), and instructions for use. [00216] In some aspects, an array or method of the invention may be used to provide a service. For example, a compound, or information identifying a compound (such as name, chemical structure, etc.) may be received from a requesting entity (e.g., electronically, such as over the Internet), together with a request to perform an assay using the compound and an inventive array. The request may further specify the nature of the assay to be performed, concentrations of compound to be used, agents to be used on the array (agents that result in expression of ORFs, agents that result in expression of shRNAs, virus arrays, plasmid arrays, categories of genes (e.g., kinases, GPCRs, etc.), etc. The assay is performed, and raw data or results (optionally with analysis thereof) are transmitted (e.g., by sending them over the Internet or by delivering a tangible representation of the results such as a paper report, computer-readable medium, etc.) to the requesting entity or its designee, or results are otherwise made available for access by the requesting entity or its designee.

[00217] VII. Exemplary Applications

[00218] This section describes a variety of uses for arrays and methods of the present invention. It is noted that the uses may be divided into different categories for purposes of convenience and that such categories do not imply any limitation on the purposes for which the invention may be used. Furthermore, uses may fall into multiple categories. It is also noted that various applications and uses are described elsewhere herein. In some aspects, the invention provides methods in which an array of the invention is employed in any use described herein.

[00219] In general, arrays of the invention can be used in any of a wide variety of applications for which cell arrays or cell-based screens are or have been used in the art. Various examples of such uses are found, e.g., in U.S. Pat. Nos. 6,544,790 or 6,951 ,757; PCT/US2002/021972 (WO/2003/056293); and/or PCT/US2006/029068 (WO/2007/014282). In general, such uses can include (a) target identification and analysis; (b) target validation; (c) screening; (d) lead optimization; (e) compound profiling. Many uses of inventive arrays include contacting an array with a test compound or test compounds by, e.g., adding such compound(s) to culture medium in which an array is maintained. In general, any test compound can be used in various embodiments of the invention. In many embodiments a test compound is a small molecule. Test compounds that are siRNAs, aptamers, polypeptides, antibodies, lipids, polysaccharides, or inorganic compounds, etc., may be used in various embodiments. In some embodiments test compounds are approved drugs or are drug candidates (compounds that are or have been under evaluation, testing, or consideration for use as therapeutic or diagnostic agents in human or non-human animal subjects). In some exemplary embodiments, a compound of interest is a kinase inhibitor, an NF-kB pathway inhibitor, a proteasome inhibitor, or a GPCR inhibitor.

[00220] Numerous small molecule kinase inhibitors that show activity, e.g., in one or more kinase inhibition assays known in the art, have been discovered. Any such kinase inhibitor can be tested using an inventive array. Exemplary kinase inhibitors and kinase targets thereof are discussed, e.g., in Zhang J, et al., Targeting cancer with small molecule kinase inhibitors. Nat Rev Cancer. 9(l):28-39, 2009; Janne PA, et al, Factors underlying sensitivity of cancers to small-molecule kinase inhibitors. Nat Rev Drug Discov. 8(9):709-23, 2009; Li, R. and Stafford, JA, Kinase Inhibitor Drugs (Wiley Series in Drug Discovery and Development), Wiley, 2009; and/or Matthews, DJ and Gerritson, M, supra.

[00221] In some embodiments, a kinase inhibitor is a compound known in the art as an inhibitor of at least one tyrosine kinase. In some embodiments, a kinase inhibitor is a compound known in the art as an inhibitor of at least one serine/threonine kinase. Kinase inhibitors have been classified into four main types. Type I inhibitors are ATP-competitive compounds that recognize and bind to the "active" conformation of the kinase, i.e., the confomiation otherwise conducive to phosphotransfer. Type II inhibitors recognize and bind to the inactive conformation of the kinase. Type III inhibitors (also termed "allosteric" inhibitors) bind outside the ATP binding site at an allosteric site. Covalent inhibitors are capable of forming a stable covalent bond to the kinase, e.g., to the kinase active site, often by reacting with a cysteine residue. In some embodiments, a kinase inhibitor is a compound known in the art as a Type I inhibitor. In some embodiments, a kinase inhibitor is a compound known in the art as a Type II inhibitor. In some embodiments, a kinase inhibitor is a compound known in the art as a Type III inhibitor. In some embodiments, a kinase inhibitor is a compound known in the art as a covalent inhibitor.

[00222] In some embodiments, a kinase inhibitor is imatinib, gefitinib, erlotinib, nilotinib, dasatinib, suniti ib, sorafenib, pazopanib, lapatinib, axitinib, brivanib, motesanib, crizotinib, torinl, torin2, AZD-6244, PD-0325901, BEZ-235, GNF-2, GNF-5, PLX4032, PLX4720, GDC-0879, PD-166326, PD-173955, PD-0332991, DV2-273, MLN8237, GSK1070916A, fostamatinib (R9355788), JNJ-26483327, GW-786034, MLN-518, MLN- 8054, VX- 680/MK-0457, PTK-787, ZD-6474, AZD1152HQPA, CHIR-258/TKI-258, AST-487, ABT- 869, riscovitine, flavopiridol, or a structural analog of any of the foregoing. [00223] Kinase inhibitors may often referred to based on the name of their primary target (and/or one or more secondary target(s)). A primary target may be one that the kinase is known to inhibit with an IC50 of 500 nM or less, e.g., 100 nM or less, in an in vitro kinase assay, and/or may be the first kinase to be recognized as a target of the kinase inhibitor and/or may be an intended target of the kinase. A secondary target may be a subsequently identified target. In some embodiments, a kinase inhibitor is a RAF inhibitor, e.g., a RAFB inhibitor. In some embodiments, a kinase inhibitor is a MEK inhibitor. In some embodiments a kinase inhibitor is an mTOR inhibitor. In some embodiments, a kinase inhibitor is a Src inhibitor. In some embodiments, a kinase inhibitor is an ABL inhibitor. In some embodiments, a kinase inhibitor is a CDK inhibitor. In some embodiments, a kinase inhibitor is a JAK inhibitor. In some embodiments, a kinase inhibitor is a KIT inhibitor. In some embodiments, a kinase inhibitor is a VEGFR inhibitor. In some embodiments, a kinase inhibitor is an EGFR inhibitor. In some embodiments, a kinase inhibitor is an ERBB2 inhibitor. In some embodiments, a kinase inhibitor is an ALK inhibitor. In some embodiments, a kinase inhibitor is a PDGFR inhibitor. In some embodiments, a kinase inhibitor is an Aurora kinase inhibitor.

[00224] A test compound can be provided in any of a variety of ways in various embodiments. In some embodiments, a test compound is added to a cell-seeded array after seeding. In some embodiments, a test compound is added to cells prior to their use to seed an array. Additional test compound may be added during culture of the array in some embodiments. For example, additional test compound may be added after 24 or 48 hours (or at intervals of 24 or 48 hours), or when culture medium is changed. In some embodiments, a test compound is provided as a component of a conditioned medium. It will be understood that a test compound may be provided as a salt of the active agent, e.g., a pharmaceutically acceptable salt, or in some embodiments as a prodrug, active metabolite, etc.

[00225] In some embodiments, a compound is tested at multiple different concentrations. For example, screens may be performed using at least 2, 5, 8, or more different concentrations, e.g., between 2 and 10 concentrations. The concentrations may vary from one another by factors of 5, 10, 20, etc. They may, for example, span between 2 and 10 orders of magnitude. Exemplary concentrations are 100 μΜ, 10 μΜ, 1 μ , 0.1 μΜ, 0.01 μΜ, 0.001 μΜ, 0.0001 μΜ and 0.00001 μΜ. In some aspects, inventive arrays make it possible to rapidly and cost-effectively determine concentration-response relationships. [00226] A "target" in the context of target identification, analysis, or validation typically refers to a cellular molecule, often a gene product (e.g., a protein), on which a compound of interest acts or whose modulation (e.g., enhancement or inhibition of expression or activity) may result in a phenotype of interest (e.g., a phenotype that may be of interest from the standpoint of drug discovery or therapy). For purposes of convenience, a gene that encodes a gene product that is a target may also be referred to as a target, and vice versa. A compound of interest may modulate the expression or activity of a target, e.g., it may enhance or inhibit the expression or activity of the target. In many embodiments a target is a cellular molecule whose modulation by a compound of interest results or is expected to result in a cell phenotype. In some embodiments the phenotype may be of interest for diagnostic, therapeutic, or research purposes. For example, a phenotype may be one that, if occurring in an organism, would be expected to be beneficial for purposes of treating a disease, or a phenotype may be one that, if occurring in an organism, would be expected to be harmful or otherwise undesirable. A target may be a "direct target", which, as used herein, means that the compound of interest physically interacts with, e.g., binds to, the target, or may be an indirect target. In the case of an indirect target, a compound of interest may bind to another cellular molecule that in turn acts directly on the target.

[00227] Target identification can include identifying a cellular molecule that is a target of a test compound or identifying a gene or gene product that may be a useful target for drug development (e.g., for identification of compounds that modulate the target and, potentially, advancement of a compound towards regulatory approval and/or acceptance in the art for administration to human or non-human animal subjects to treat or diagnose a disease). Assays useful for target identification can include binding assays. For example, if a test compound binds more strongly to cells that overexpress a particular protein as compared with binding to cells that do not overexpress the protein, the protein may be a target of the test compound. In some embodiments, target identification can be performed using RNAi or overexpression approaches. For example, if inhibiting a gene results in a potentially useful phenotype, then the gene may be an appropriate target for drug development (e.g., for development of compounds that inhibit an expression product of the gene). As another example, if inhibiting expression of a particular gene by RNAi alters the effect of a test compound on a cell, the gene may be identified as a target of the compound. If increasing the amount of a particular gene product by overexpressing the gene in a cell alters the effect of a test compound on the cell, the gene product may be a target of the test compound. Target identification can include determining that a suspected or candidate target of a compound of interest is or is not a target of such compound.

[00228] Target validation can include verifying that the inhibition or activation of a particular gene product has or does not have a particular effect of interest on cell phenotype. For example, if inhibiting expression of a particular protein in cells (e.g., by expressing an RNAi agent in the cells) results in a particular effect on cell phenotype (e.g., an effect of potential therapeutic benefit), then the protein may be considered to be a suitable target for the development of drags that act on (e.g., inhibit) that protein. If inhibiting expression of a particular protein in cells results in an undesirable phenotype (e.g., a potentially deleterious phenotype likely to result in unacceptability for administration for therapeutic purposes), then the protein may be considered not to be appropriate as a target for drug development, or it may be necessary to exercise appropriate care in patient selection.

[00229] In some aspects, a potential drug candidate may be evaluated for selectivity by incubating the candidate with a cell array comprising agents that cause cells to express potential targets. An array could represent the entire set of genes in the genome(s) of interest or focused subsets, e.g., kinases, GPCRs, ion channels, enzymes, nuclear hormone receptors. The relative binding of the drug candidate to one or more known targets and/or other potential targets could be determined or the ability of the drug candidate to modulate the activity of one or more known targets and/or other potential targets could be determined. Candidates with a high degree of non-selective binding or non-selective activity could be abandoned or modified to reduce non-selective binding or non-selective activity before additional testing such as ADME or toxicology or other tests (e.g., prior to performing tests involving use of non-human animals). Test compounds, e.g., potential drug candidates, could be evaluated for toxicity by incubating the candidate with the appropriate array of targets, such as cytochrome P-450s, including pharmacogenomic variants or other variations. In some embodiments, it may be desirable that a compound affects multiple different targets. For example, a number of kinase inhibitors may exert therapeutic effects by inhibiting multiple kinases. Clinical resistance to kinase inhibitors can arise from mutations, and it may therefore be desirable to identify compounds capable of inhibiting one or more mutant forms of a kinase. Compounds having a desired spectrum of activity (e.g., for multiple different kinases or multiple mutant forms of a kinase) may be identified using arrays of the invention.

[002301 Selectivity tests could be performed on the metabolites of a test compound, e.g., a potential drug candidate. For instance, a radiolabeled compound could be reacted with a biotransformation agent, such as a liver extract or fraction thereof (e.g., liver microsomes), tissue culture system, or living organism such as a rodent or dog. The radiolabeled metabolites could then be extracted and purified and tested for binding with the array or for biological activity. Metabolites with binding activity or biological activity could then be characterized further by standard methods.

|00231] In some embodiments, a screen is performed to identify genes and gene products thereof whose expression renders cells less sensitive (or, equivalently, more resistant) to a compound of interest that is capable of causing a phenotypic change in the cells. A first cell or cell population is considered to be "less sensitive" or "more resistant" to a compound than a second cell or cell population if the first cell exhibits a lesser phenotypic change in the presence of the compound than does the second cell or cell population under the same conditions (e.g., when exposed to the same concentration of the compound for the same length of time). In some embodiments a compound of interest inhibits survival or proliferation of at least some mammalian cells (i.e., the phenotypic change caused by the compound is a reduction in survival or proliferation). In some embodiments a compound of interest inhibits survival or proliferation of at least some types of tumor cells. In some embodiments, the cells are tumor cells, e.g., any of the tumor cell types mentioned above. In some embodiments, the cells, e.g., tumor cells have a mutation in a gene that encodes a target of a compound, wherein the mutation renders the cells less sensitive to a compound of interest than comparable cells (e.g., genetically matched cells) that do not have the mutation. In many embodiments the mutation alters the sequence of a protein that is a target of the compound. The mutation may be naturally occurring or genetically engineered in various embodiments. A screen may identify a protein whose expression renders tumor cells less sensitive to an inhibitor of cell survival or proliferation. Such a protein (or other protein(s) in the pathway in which that protein functions) would be appropri te targets for development of therapeutic strategies to reduce emergence of resistance or counteract resistance to the compound (or to other compounds that act on the same target). For example, a compound that inhibits expression or activity of a protein whose expression promotes resistance to a compound of interest may be useful to reduce emergence of resistance or counteract resistance to the compound of interest.

[00232] The identification of proteins whose expression or overexpression promotes resistance may be useful for predictive purposes relating to therapy, e.g., for tumor treatment. "Predictive purposes", can include, for example, predicting the likelihood that a cell, organ, tissue, or abnormal growth is sensitive to a compound or the approximate extent of such sensitivity, predicting whether a subject suffering from a disease is likely to experience a partial or complete response to the compound, determining whether a subject is an appropriate candidate for treatment with a compound, determining whether a treatment is an appropriate treatment for a subject, or selecting a treatment for a subject. "Predictive purposes", in the context of cancer therapy, can include, for example, predicting the likelihood that a tumor is sensitive to a compound or the approximate extent of such sensitivity, predicting whether a subject suffering from a tumor is likely to experience a partial or complete response to the compound, determining whether a subject is an appropriate candidate for treatment with a compound, determining whether a treatment is an appropriate treatment for a subject, or selecting a treatment for a subject. A biological sample, e.g., a tumor sample, can be tested to determine whether it comprises cells that express or overexpress a protein that promotes resistance to a potential therapeutic agent. If the tumor comprises such cells, a different therapeutic agent may be selected, or a combination therapy that includes an inhibitor of the resistance promoting protein can be selected. If the tumor does not comprise such cells, the potential therapeutic agent may be selected. A biological sample can be any sample obtained from a subject. Typically, a biological sample (as obtained) comprises one or more cells (e.g., tumor cells). In some embodiments, the sample comprises a tissue or cell sample, e.g., a surgical biopsy sample, a fine needle biopsy sample, cell brushing or washing, blood, etc. It will be understood that the sample can be processed or subjected to any of a variety of procedures after having been obtained, e.g., for purposes of facilitating analysis thereof.

[00233] In some embodiments agents on an array can represent a wide variety of potential genotypic or phenotypic variations (e.g., among humans), such as a wide variety of polymorphic variants of drug metabolizing enzymes (e.g., CYPs) or a range of different expression levels of multiple different proteins. Cell arrays comprising cell features expressing such variants can be can be contacted with one or more test compound(s), and polymorphic variants or expression levels that are associated with particular responses to the test compound(s) (e.g., increased or decreased production of metabolites), alterations that may be indicative of potential toxicity or potential therapeutic efficacy, can be identified. Genotypes or phenotypes associated with (e.g., correlating with) one or more such responses may be identified. Such methods may be useful in drug development efforts by, for example, facilitating identification of subpopulations that are more or less likely to benefit from, or to suffer side effects from, a particular compound under development or being considered for development as a therapeutic agent. In some aspects, arrays of the invention may find use in personalized medicine applications, wherein, for example, an individual having a genotype or phenotype identified using a cell array may be identified as being a suitable or unsuitable candidate for treatment with the compound, or the compound may be identified as appropriate or not appropriate for administration to the individual.

[00234] In some aspects, the invention finds use in compound (e.g., small molecule) profiling. Use of arrays of the invention allows, for example, the rapid screening of thousands of compound/ORF or compound/loss of function combinations. Such screens may be used, for example, to identify genes whose overexpression or loss of function may contribute to drug resistance (response modifiers), drug efficacy, or unwanted side effects. In some aspects, the invention provides a variety of screening assays. In a broad sense, "screening" can include any use of an array in which a test compound or agent having a selected effect (e.g., a potentially therapeutically useful effect) on cell phenotype is sought. Screening often includes assessing the effect of many (e.g., hundreds, thousands, or millions) of distinct test compounds, agents, or test compound/agent combinations on one or more cell phenotypes of interest. In some embodiments, a cell phenotype of interest is a "response" to a compound. A response can be, e.g., an increase or decrease in cell viability or cell proliferation, an alteration in one or more biological functions or processes of the cell, an alteration in expression or activity or subcellular localization or post-translational modification of one or more gene products, etc. A cell that exhibits a particular response of interest when contacted with a compound may be said to "respond" to the compound or to be "sensitive" to the compound. A cell that does not exhibit the response or exhibits a reduced response as compared, for example, with a sensitive cell may be said to be "resistant" to the compound. In many embodiments a cell response of interest in a culture environment (ex vivo) may correspond to or correlate with a response of interest in vivo (i.e., in a human or animal). For example, a reduction in cancer cell viability or proliferation in culture in response to a compound may correlate with reduction in cancer cell viability or proliferation in vivo and may result in therapeutic efficacy in a subject with cancer. In some embodiments a screen is performed to identify compound response modifiers, e.g., genes whose overexpression or inhibition modifies cellular response to a compound of interest. In some embodiments a screen is used to identify useful compound combinations or targets that would be useful to modulate (e.g., inhibit) in combination. A "combination therapy" typically refers to administration of two or more compounds sufficiently close together in time to achieve a biological effect (typically a therapeutically beneficial effect on a particular disease or condition) which is greater than or more beneficial or more prolonged than that which would be achieved if any of the compounds were administered at the same dose as a single agent or that would be useful to maintain efficacy (e.g., by inhibiting emergence of drug resistance). In some embodiments two or more compounds are administered at least once within 6 weeks or less of one another. Often, the two or more compounds may be administered within 24 or 48 hours of each other, or within up to 1, 2, 3, or 4 weeks of one another. In some embodiments they may be administered together in a single composition but often they would be administered separately and may be administered using different routes of administration or the same route of administration. Combination therapy may, for example, result in increased efficacy or permit use of lower doses of compounds, which can reduce side effects. Compounds used in a combination therapy may target the same target or pathway or may target different targets or pathways.

[00235] In some embodiments a screen may be performed using multiple different cell types, e.g., 2-50 different cell types, or more. For example, a panel of cancer cell lines can be tested including, for example, cancer cell lines derived from lung, breast, prostate, skin, brain, pancreas, colon, kidney, liver, and/or other tissues. In some embodiments, screens are performed using multiple cancer cell lines of a given cancer type. In some embodiments a screen may be performed using a cell type that may be of particular relevance with regard to a phenotype of interest, such as cells of a cell type that is affected in a disease for which a drug candidate or target is sought or that may be particularly vulnerable to an undesired side effect of a compound. In some embodiments a screen is used to identity or assess genetic interactions such as suppressive or enhancement effects. Examples of such effects include, e.g., "synthetic lethality" or "synthetic viability" interactions. Synthetic lethality refers to a genetic interaction in which single gene defects or effects (e.g., loss of function or overexpression) are compatible with cell viability but a combination of two (or more) such genetic defects or effects (e.g., results in cell death (or significant impairment of fitness, sometimes termed "synthetic sickness"). Synthetic viability refers to a situation where a combination of gene effects or defects rescues the lethal effects of a single gene effect or defect. In some embodiments a screen may be performed to identify or characterize "gene addiction", e.g., which refers to a situation in which a cell has become completely dependent for viability upon the activity of a gene (e.g., a mutated or aberrantly expressed gene). In cancer, tumor cells may exhibit oncogene addiction. Gene products of genes to which tumor cells become "addicted" may be attractive drug targets.

[00236] In some embodiments, a screen may be performed to identify gene products that render mutant cancer cells (e.g., cancer cells harboring a mutation in a kinase that is a target for a kinase inhibitor) less sensitive to an inhibitor of said kinase. For example, an array comprising cells (e.g., mutant cancer cells) that have been transfected with agents causing them to overexpress a set of human ORFs is provided. The array is contacted with a kinase inhibitor that would (in the absence of a response modifier) inhibit survival or proliferation of the cells. Cell spots in which the survival or proliferation of the cells is significantly greater as compared, for example, with cell spots on control regions lacking an agent containing a human ORF are identified, thus identifying ORFs (and their encoded gene products) whose expression confers resistance to the kinase inhibitor. Such gene products may be potential targets for development of combination therapies and/or are useful for prediction of drug sensitivity or resistance. The set of human ORFs may be focused (e.g., kinase ORFs) or may be a diverse set of human ORFs (e.g., representing a substantial fraction of the genome). The identified ORFs may function in one or more pathways that may be identified, for example, using a method such as gene set enrichment analysis. This approach may be used to identify previously unrecognized pathways for potential combination therapy development or drug sensitivity prediction.In some aspects, arrays are used to identify one or more compounds that have similar activity and/or specificity to that of a selected compound of interest and/or that differ in one or more ways with regard to activity and/or specificity as compared with a selected compound of interest. For example, a compound of interest can be profiled against a cell array of the invention. One or more test compounds are profiled against a substantially identical cell array of the invention (e.g., an array comprising the same set of agents). The resulting profiles are compared to identify compound(s) that have similar activity and/or specificity as the compound of interest. For example, compound(s) that have (i) increased activity towards one or more target(s), wherein activity towards the target is expected to be or has been determined to be beneficial; (ii) reduced activity toward one or more targets, wherein activity towards the target has been determined to be or expected to be deleterious; and/or (iii) increased or decreased selectivity, may be identified.

[00237] In some aspects, inventive arrays are used to profile multiple (e.g., tens, hundreds, or thousands) of compounds across multiple ORFs or shRNAs with regard to effect on one or more phenotypes of interest (e.g., cell survival or proliferation or activation state of a signaling pathway). For example, the effect of each of the ORFs or shRNAs on cell response to the compound is evaluated. Assays could be performed across multiple cell lines (e.g., multiple cancer cell lines). Assays could be performed using compounds individually or in combinations. In some embodiments at least some of the compounds may be accepted in the art as having a particular biological effect or as being useful for, e.g., therapeutic or research purposes. In some embodiments a database containing the results is generated. The database may be stored on a computer-readable medium. Subsequently, a compound that may be relatively uncharacterized may be profiled across at least some of the same ORFs or shRNAs, typically across most or all of these ORFs or shRNAs. The results are compared with those obtained with the previously tested compounds. In many embodiments, the comparison would be performed using appropriate computer software to, for example, identify compounds having similar profiles. Any of a variety of approaches may be used to analyze the data. Results may, for example, help to identify a target of the compound or a pathway on which the compound acts, may suggest a utility for the compound, etc. In some embodiments, a compound that is relatively well characterized and may be, for example, an approved drug, can be profiled. Results could, for example, potentially suggest previously unrecognized potential therapeutic uses for the drug or potentially useful combination therapies that include the drug. This approach may be valuable, for example, in the case of compounds that failed to meet efficacy endpoints in one or more clinical trials. Alternative indications or criteria for subject enrollment for future clinical trials may be identified or useful combination approaches may be identified. In some embodiments the database could be queried by a requestor to identify a compound having a desired profile or characteristics. Optionally, a collection of tested compounds is maintained, and a sample is provided to the requestor.

[00238] VIII. Methods of Inhibiting or Predicting Cell Survival. Proliferation, or Drug Resistance

[00239] In some aspects, the invention provides methods of inhibiting tumor cell survival, proliferation, or drug resistance. The methods are based at least in part on insights gained from screens performed using arrays of the invention (see Examples). As described in further detail in the Examples, the NF-κΒ pathway was identified as a mediator of resistance to MAPK inhibitors. NF-κΒ pathway activity may thus predict clinical efficacy of MAPK inhibitors, e.g., for treatment of cancer. In one aspect, the invention provides a method of determining whether a tumor cell, tumor cell line, or tumor will be sensitive to a MAPK inhibitor, the method comprising measuring NF-κΒ pathway activity of the tumor cell, tumor cell line, or tumor, wherein if NF-κΒ pathway activity is increased relative to a control, then the tumor cell, tumor cell line, or tumor has an increased likelihood of being resistant to the MAPK inhibitor (or, equivalently, the tumor cell, tumor cell line, or tumor has a decreased likelihood of being sensitive to the MAPK inhibitor) as compared with the likelihood if NF- KB pathway activity is not increased. NF-κΒ pathway activity can be measured using, e.g., a suitable reporter gene construct, an NF- Β pathway activity transcriptional signature comprising at least one gene whose expression is regulated by the NF-κΒ pathway, etc, (See Examples for exemplary NF-KB-regulated genes and NF-κΒ transcriptional signature.) A suitable control value can be, e.g., an average value measured in a panel of cell lines that are sensitive to a MAPK inhibitor (e.g., UACC-62, Sk-Mel-5, Sk-Mel-28, Malem-3M, Colo679, A375). If a tumor is deemed likely to be resistant to the MAPK inhibitor, appropriate action may be taken. For example, a different therapeutic approach may be selected, or, in some embodiments, an NF-kB inhibitor may be administered in combination with the MAPK inhibitor. On the other hand, if a tumor is deemed likely to be sensitive to the MAPK inhibitor (e.g., the tumor has low levels of NF-kB pathway activity), a MAPK inhibitor may be selected as a therapy. In some embodiments, the invention provides a method of inhibiting survival or proliferation of a tumor or tumor cell comprising contacting the tumor or tumor cell with an NF-κΒ inhibitor and a MAPK inhibitor. The invention contemplates, in various embodiments, the combination of any MAPK inhibitor with any NF-kB inhibitor. The invention contemplates, in various embodiments, use of the combination to inhibit tumor cell growth or survival, e.g., in cell culture or in vivo, e.g., for treatment of cancer. The invention contemplates administering an NF-kB inhibitor in combination with a MAPK inhibitor to reduce the likelihood of emergence of resistance to the MAPK inhibitor. In some embodiments, the invention contemplates the combination of a MAPK pathway inhibitor; and (ii) a Src pathway inhibitor, mTOR pathway inhibitor, pI3 kinase pathway inhibitor, or protein kinase C (PKC) pathway inhibitor. The invention contemplates, in various embodiments, use of such combinations to inhibit tumor cell growth or survival, e.g., in cell culture or in vivo, e.g., for treatment of cancer. In some embodiments, a combination therapy or predictive method of the invention is applied in the context of a tumor that has an activating mutation or activating chromosomal translocation or amplification in a MAPK pathway component (e.g., RafB). [00240] The MAPK pathway (also known as the Ras/Raf/MEK/ERK pathway or cascade) is well known in the art. "MAPK inhibitor" or "MAPK pathway inhibitor" as used herein can be any compound that inhibits one or more proteins in the MAPK pathway. For example, a MAPK inhibitor may inhibit Ras, Raf, or one or more MEKs (e.g., MEK1 or MEK2), or in some embodiments may inhibit a cell surface receptor that activates the MAPK pathway. MAPK pathway inhibitors are useful for treatment of a variety of cancers, e.g., cancers that have mutations in one or more MAPK pathway components. In some embodiments a MAPK pathway inhibitor inhibits Raf kinase (e.g., RAFB). Exemplary Raf kinase inhibitors include, e.g., sorafenib, SB590885, PLX4720, XL281, RAF265, and vemurafenib. In some embodiments a MAPK pathway inhibitor inhibits a MEK kinase (e.g., MEK1 and/or MEK2). Exemplary MEK kinase inhibitors include, e.g., selumetinib, XL518, CI- 1040, PD035901, and GSK1 120212. In some embodiments a MAPK pathway inhibitor is relatively specific for a particular target, while in some embodiments a MAPK pathway inhibitor inhibits multiple targets within the MAPK pathway, or including one or more targets not in the MAPK pathway. See, e.g., Chappell, W., et al., Oncotarget, March 201 1 , Vol.2, No 3, 135 - 164, for discussion of exemplary MAPK inhibitors, mTOR pathway inhibitors, and other kinase pathway inhibitors, including a number of compounds that are in human clinical trials or approved for treatment of cancer, which may be used in embodiments of the invention.

[00241] An "NF-kB inhibitor" or "NF-kB pathway inhibitor" as used herein can be any compound that inhibits the synthesis, activation, translocation, and/or DNA binding activity of NF-kB or otherwise affects NF-kB activity. Such modulation can often involve modulating one or more proteins upstream of NF-kB in the NF-kB pathway (e.g., a component of an NF-kB signaling module). An NF- inhibitor could inhibit NF-KB by intervening in the NF-kB signaling pathway in any of a variety of ways, in various embodiments of the invention. In some embodiments, an NF- inhibitor may inhibit nuclear translocation of NF-kB. In some embodiments, an NF- inhibitor inhibits phosphorylation and degradation of IkBa or reduces NF-kB activation through the formation of conjugates with NF-kB.

[00242] In some embodiments, an NF-kB inhibitor is a small molecule, antioxidant, peptide, small RNA or DNA, microbial or viral protein, or engineered dominant-negative or constitutively active polypeptides or peptide, decoy oligonucleotide, etc. In some embodiments, a NF-kB inhibitor comprises an RNAi agent (e.g., an siRNA) or an antisense oligonucleotide that inhibits expression of one or more NF-kB genes (e.g., genes encoding NFKB1, NFKB2, RELA, REL, and/or RELB) or NF-kB pathway genes that contribute to NF-kB activity. In some embodiments an NF-kB inhibitor comprises an antibody or aptamer that specifically binds to an NF-kB protein (NFKB1, NFKB2, RELA, REL, and/or RELB.

[00243] Exemplary compounds that may be used as NF-kB inhibitors include 2-chloro-N- [3,5-di(trifluoromethyl)phenyl]-4-(trifluoromethyl)pyrimidine-5-carboxamide (also known as SP- 100030); 3,4-dihydro-4,4-dimethyl-2H-l ,2-benzoselenazine (also known as BXT- 51072); declopramide (also known as Oxi-104); and dexlipotam. In one embodiment, an NF- kB inhibitor is denosurnab, which inhibits RANKL, which, in turn, through its receptor RANK inhibits NF-kB. In other embodiments, an NF-kB inhibitor is a chalcone or derivative or analog thereof such as 3-hydroxy-4,3',4',5'-tetraniethoxychalcone (Srinivasan B, et al, J Med Chem. 52(22):7228-35, 2009). In other embodiments, an NF-kB inhibitor is a lignan (manassantins, (+)-saucernetin, (-)-saucerneol methyl ether), sesquiterpene (costunolide, parthenolide, celastrol, celaphanol A), diterpenes (excisanin, kamebakaurin), triterpene (avicin, oleandrin), polyphenol (resveratrol, epigallocatechin gallate, quercetin), or a derivative or analog thereof (Mini Rev Med Chem. 6(8):945-51, 2006), In some embodiments, an NF-kB inhibitor inhibits NF-kappaB signaling at least in part via inhibition of IkappaBalpha phosphorylation. In some embodiments, such an NF-kB inhibitor is emetine, fluorosalan, sunitinib malate, bithionol, narasin, tribromsalan, or lestaurtinib (Miller SC. Biochem Pharmacol. 79(9): 1272-80, 2010), or a derivative or analog of any of these. In some embodiments, an NF-kB inhibitor is an anti-oxidant that has been shown to inhibit activation of NF-kB, proteasome or protease inhibitors that inhibits Rel/NF-kB, or an IkBa phosphorylation and/or degradation inhibitor. A variety of compounds reported to inhibit one or more steps of NF-kappaB signaling are described in, e.g., Qilmore TD, & Herscovitch M. Oncogene, 25(51):6887-99, 2006 and/or at the following website

http://people.bu.edu gilmore/nf-kb/inhibitors/index.htmK incorporated herein by reference as of September 15, 2011. In some embodiments, a compound of use is a non-steroidal antiinflammatory agent (e.g., sodium salicylate, 5-aminosalicylic acid, ibuprofen, sulindac, indomethacin), BAY-II, thalidomide, flavopiridol, PS-341 (bortezomib), Silibinin, Leptomycin B, Sesquiterpene lactones (parthenolide, ergolide), derivative of 9 aminoacridine (9AA): antimalarial agent, celecoxib (celebrex; Pfizer) COX2 inhibitor, dimethylamino- parthenolide (DMAPT), or diethylmaleate. Exemplary Src pathway inhibitors include, e.g., AZD0530, bosutinib, dasatinib, KX2-391, and XL-999. Examples

[00244] Example 1: Development of the MicroSCALE screening platform

[00245] We designed a screening platform in which lentiviruses encoding ORFs or shR As are printed as small (300-600 μηι diameter) features on glass slides which, when seeded with adherent cells, yield spatially-defined islands of infection that can be assayed for phenotypes of interest (Fig. la). This design was realized through a series of technical developments.

[00246] First, we created a methodology to spatially confine infected cell populations into discrete, densely-spaced, regular islands that can be assayed for phenotypes of interest in a manner analogous to multiwell plates. Spatial confinement of cells on glass slides was problematic as the size and localization of the region of infection adjacent to each printed feature varies substantially from cell line to cell line due to cell migration away from printed features, leading to irregular and poorly localized regions of infection (Supplementary Fig. la). {To achieve uniform spatial confinement of cells to printed microarray features, we first coated a glass slide with a material that is resistant to cell adhesion and then printed features in two sequential steps: first with a material that promotes localized cell adhesion, then with individual lentiviruses (directly on top of the printed adhesive regions) (Supplementary Fig. lb). The resultant surfaces produced robust cell adhesion and infection only within printed areas, resulting in islands of uniform size and shape regardless of the cell line used (Fig. lb and Supplementary Fig. lc). In our studies, we have used polyacrylamide hydrogel-coated surfaces as the adhesion-resistant substrate coating and gelatin as the printed cell adhesive material, although other materials could be used, such as those described in reference 12 and/or above.

[00247] Second, we developed a method to achieve consistent and efficient infection of cells adhering to printed lentiviral features. This was essential for all downstream screening applications and requires (1) effective binding of viruses to the array surface and (2) the use of highly concentrated (i.e. high titer) virus preparations. Conventional lentiviral preparations were found not to permit desired le vels of virus-surface binding due to the presence of serum proteins in viral harvest media (Supplementary Fig. 2a). Moreover, the titer of unconcentrated lentiviral preparations (~ 5 x 10⁷ IFU/mL) is inadequate to achieve high rates of infection on printed features (Supplementary Fig. 2b). Current techniques for lentiviral concentration and purification typically rely on high speed centrifugation of a small number of samples (<10), a process that is incompatible with or, at best, poorly suited for the parallel, high-throughput concentration of the hundreds to thousands of distinct lentiviruses we aimed to employ in our platform¹³. Accordingly, we devised a dual purification and concentration technique that permits high-throughput, parallel preparation and subsequent printing of hundreds of unique high titer viruses in a single day. This "polyelectrolyte complexation" method, adapted from a technique originally described by Le Doux et al.¹⁴, involves the sequential addition of oppositely charged polyelectrolytes to lentiviral supernatants to form a polymer complex that entraps lentiviruses by electrostatic interactions. The polymer-virus complex can then be pelleted by low speed centrifugation and mechanically resuspended in a desired volume of printing buffer to yield a concentrated virus solution ready for printing (Supplementary Fig. 3a), This purification/concentration method is highly robust, can be performed in multiwell plates using a conventional benchtop centrifuge at low speed, and yields approximately 80% recovery of functional virus particles (Supplementary Fig. 3b). Once printed, concentrated lentiviruses adhere well to the underlying surface and efficiently infect cells on icroSCALE features. Further, because all ORF- and shR A-expressing lentiviruses in our libraries contain mammalian selection markers (blasticidin and puromycin, respectively^6, ''), we can apply selection to ensure that all cells assayed on each feature are infected (Fig lc and Supplementary Fig 4). In addition, we found that the infectivity of concentrated, printed lentiviruses is not affected by long-term storage up to at least 8 months at -80°C; data not shown).

[00248] To test the generality of the MicroSCALE platform using the optimized slide coating and printing processes, we seeded microarrays with a collection of mammalian cell lines (including cancer cell lines derived from lung, breast, prostate, skin, brain, pancreas, colon, kidney, liver, and other tissues), genetically engineered, transformed human mammary epithelial cells, and mouse embryonic fibroblasts. Of 37 cell lines tested to date, 30 (81%) exhibited efficient adhesion, growth, infection, and selection on MicroSCALEs and are thus deemed suitable for high-throughput screening, underscoring the broad potential utility of this platform (Supplementary Table 2).

[00249] In addition, we developed a simplified approach to quantify MicroSCALE screening results. Automated microscopy, the standard method of analyzing cells on glass slides, can capture complex cellular phenotypes but requires specialized equipment and takes hours to days to analyze the hundreds of features on a single slide¹⁶, making it a potentially rate-limiting step for large screens. We found that certain simple, commonly measured phenotypes such as cell number and phosphoprotein levels do not require cellular-level imaging and can instead be accurately quantified on circular MicroSCALE features by measuring only the total fluorescence intensity on each feature using a conventional DNA microarray scanner (Supplementary Fig. 5). For example, the total intensity of a nuclear

DNA stain on a feature is linearly proportional to the number of cells present on that feature

(Fig S5a). Thus, in applicable screens, this method can be used to analyze an entire slide in minutes, providing a means for rapid analysis of large quantities of MicroSCALE screening data. As the time that is required to scan, image, and process an array is approximately 5 min, this advance can translate into a one to two orders-of-magnitude decrease in the time required to acquire and analyze screening data (Table 1 and Supplementary Table 1).

Table 1 : Materials requirements and throughput estimates for kinome-scale drug modifier screens in 96-well plate and MicroSCALE formats*

MicroSCALE 96-well plate

Maximum throughput 25-50 < 5

(screens / week / scientist)

Cells required / screen 0.2-0.4 x 10⁶ 4-8 x 10⁶

Media required / screen (mL) 50-100 1 ,000-1 ,500

Drug required / screen 20-40 400-800

(mL, at screening concentration)

Detection reagent required / screen (mL) < 5 50-100 (ex: viability stain or antibody)

Number of arrays or plates required / screen 4 (arrays) 40 (plates)

"•"Calculations are based on the assumption of triplicate measurements per ORF per condition (drug/vehicle) for 96-well plate screens and sextuplicate measurements per ORF per condition for MicroSCALE screens, both of which are widely considered as standard. Costs of MicroSCALE screens are between one and two orders-of-magnitude lower than 96-well plate screens depending on parameters such as the number of replicates used and the choice of detection reagent.

pooled, and MicroSCALE screens

«itli

Supplementary Table 2: Examples of cell lines compatible with Mfcrt>3CALE screening*

A5 9 293T S 480

HeLa Pane-: HCT-Ι Ιβ

78S-0 S 620 HCC-827

HepG2 HBL-100 A375

U20S CF-7 DOLS8817

DU1 5 PCS LPS141

90-8T DA-MB-231 Sk-Br-3

U87 MDA-MB-4S3 UACC-812 EF Various human Z -75-1

mammary epithelial cell

derivatives

^*Ce! lines were (teemed compatible writh lcroSCALE screening t they

erfiitttedl good attachment, grow*, infection, and drug selection on

mtaoarreye. In total; 3W37 tested tines {81 %) paeaad el of these criteria aid were thus deemed suitable for screening.

[00250] Example 2: Selected Applications of the MicroSC'ALE screening platform

[00251] To investigate whether the MlcroSCALE platform could reproduce a representative set of proliferation- or staining-based results found by standard multiwell plate-based assays, we performed a series of experiments. First, we verified that proliferation on microscopic features can be accurately monitored over time by comparing the effects of control (shGFP) and cytostatic (PLKl - and CSNK IE-targeted) shRNAs which are known to block mitotic progression (Fig. 2a)³. Second, we reproduced recent observations concerning oncogene addiction and synthetic lethality, demonstrating the dependence of £Gi¾-mutant non-small cell lung cancers on growth signaling through EGFR and ErbB3 and the dependence of KRAS-mutmt cancers on both KRAS and TBK1 (Fig. 2b)⁴'¹⁷. Third, by immunofluorescence, we demonstrated that the phosphorylation of ribosomal protein S6 requires the upstream activity of mTOR and Raptor, components of the mTORCl complex (Fig. 2c)¹⁸. [00252] We demonstrated that MicroSCALE can systematically identify drug sensitivity modifiers in a large-scale screen. We focused specifically on kinases because they mediate the activity of diverse cell signaling pathways and may be a rich source of druggable sensitivity modifiers. We applied the semi-automated virus concentration and printing techniques described above to produce Kinome ORF MicroSCALEs, each printed with 1632 features consisting of the 618 lentivirally delivered ORF constructs printed in duplicate or triplicate (Fig. 2d). The sequence-validated wild type ORFs were obtained from a previously-published collection and represent over 75% of annotated kinases as well as a curated collection of mutant alleles and well-annotated non-kinase ORFs (Supplementary Fig. 6 and Supplementary Table 3)¹⁵. Over 90% of printed constructs yielded stable infection as determined by resistance to blasticidin selection, and spot-spot cross-contamination was minimal: replicate spots containing the same ORF consistently yielded similar levels of infection and control spots containing no virus yielded no surviving cells following selection.

[00253] Using Kinome ORF MicroSCALEs, we performed a screen to identify kinases that render B-RAF^vmE mutant melanoma cells (A375) less sensitive to PLX4720, a selective RAF inhibitor^20"22 (Supplementary Fig. 7). We found excellent agreement between the results of this screen and another screen, which utilized the same cell line, drug, and a largely overlapping ORF library, but was performed in multiwell plate format with an independent viral preparation and different infection assay procedures (Fig. 2e-f)¹⁵. Of the top 11 wild type ORF hits from the original screen that gave efficient infection in our screen, 9 scored with statistical significance (p < 0.05), 8 scored in the top 10%, and 6 scored in the top 4% (top 20 ORFs). Hits included MAP3Ks C-RAF, COT (which was excluded from further analysis because of poor infection), and MOS, which scored weakly in the original screen¹⁵. Several MAPK pathway mutant alleles not present in the original multiwell plate screen were added to our screen as positive controls, including H-RA5f"^2v, MEKl^pmL, and MEK1^Q56P. Each of these also scored as top hits, as expected^15,23"25. ΡΩΟΡΚβ and IGF1R, two kinases previously implicated in RAF inhibitor resistance^26"27, did not score, possibly owing to insufficient levels of ligand or accessory factors required to achieve full activation of these receptors.

[00254] Example 2: Systematic screens enable a kinome-wide view of genetic modifiers to targeted inhibitors

[00255] To exploit the scalability of our miniaturized screening approach, we designed a strategy to integrate the results of multiple functional modifier screens across related classes of drugs with existing cell line pharmacogenomic data to nominate high-priority candidate resistance genes and pathways (Fig 3a).

[00256] Functional screens were performed as described above using five additional small molecule inhibitors relevant to melanoma, including agents targeting RAF (GDC-0879), MEKl/2 (AZD-6244 and PD-0325901), and mTOR/PI3K (BEZ-235 and Torinl)

(Supplementary Fig. 8; 2-5 replicate arrays (screens) were used for each drug along with 4 replicate vehicle-treated arrays)^28"32. Unsupervised hierarchical clustering of both arrays and ORFs revealed several notable phenomena (Fig. 3b). First, kinome-wide modifier profiles segregate the three classes of inhibitors based on their biochemical targets, with replicate arrays frequently clustering together as nearest neighbors. Second, most strong hits against drugs targeting one member of the MAP pathway (MEKl/2 or RAF) are also hits for drugs targeting the other member. On the other hand, there were a few notable exceptions to this trend, including C-RAF and MOS, which conferred selective resistance to RAF inhibitors but not MEKl/2 inhibitors. This is expected given that both C-RAF and MOS directly activate MEK^15,33. Conversely, hits rarely scored for both mTOR and MAPK inhibitors, implying that genetic modifier profiles contain functional information that is specific to the biochemical pathway(s) perturbed by each drug. Thus, while the overexpression of individual kinases can commonly confer resistance to multiple inhibitors targeting a single pathway, individual kinases can only rarely confer cross-resistance to inhibitors targeting distinct pathways.

[00257] To validate and prioritize candidates emerging from primary screens, we selected a panel of 30-40 ORFs that scored in the top 15% in individual screens with statistical significance (p < 0.05). We expressed each of these ORFs individually in A375 cells in a traditional multiwell plate format and measured proliferation scores (drug/vehicle) at a single drug concentration. Of the hits identified in primary screens, 74%, 65%, and 63% were validated as conferring a survival advantage relative to a neutral control (MEK1) in the presence of RAF, MEKl/2, and mTOR PI3K inhibitors, respectively. In general, genes that modulate drug sensitivity segregated into defined classes and pathways. For example, hits modulating sensitivity to mTOR PI3K inhibitors included protein kinase A (PKA) subunits (PRKAR1A, PRKACB, and PRKACG) and established PI3K-mTOR pathway-associated genes (RPS6KA5, PIK3CG, and PIP5K3), while those modulating sensitivity to both RAF and MEKl/2 inhibitors included multiple SRC-family kinases LCK, HCK, and FOR) and novel protein kinase C (PKC) isozymes (PRKCE, PRKCQ, and PRKCH). We also found that several NF-κΒ pathway members scored in primary screens against RAF and ME 1/2 inhibitors. Two of these genes, IKBKB (a subunit of the ΙκΒ kinase complex) and TRAF2 (a TNF receptor associated protein), were tested and validated in secondary assays

(Supplementary Fig. 9 and Fig. 3c).

[00258] Example 3: Integrating functional screens with large-scale

pharmacogenomics data using MicroSCALES

[00259] Existing pharmacogenomic data sets provide an independent means of identifying genes and pathways associated with resistance that can be used to complement and prioritize the findings emerging from large-scale functional screens. Using a panel of 25 B-RAF^vmE mutant melanoma cell lines for which steady state gene expression (mRNA) and pharmacological sensitivity data are available (see Methods)'³, we first identified cell lines that are sensitive or resistant to RAF and MEK1/2 inhibitors (Supplementary Fig 10). To determine if resistant cell lines shared common transcriptional signatures, we performed single-sample gene set enrichment analysis⁴'³⁴ (GSEA) to identify gene sets whose pattern of expression across the entire panel of cell lines most strongly correlated with the observed pattern of MAPK inhibitor resistance. Notably, we found a highly significant enrichment of multiple independent gene sets associated with NF-κΒ pathway activation in resistant cell lines compared with sensitive lines (Fig. 3d), consistent with the results of our large-scale functional screens, which found that the overexpression of NF- Β pathway genes can confer selective resistance to MAPK inhibitors (Supplementary Fig. 11).

[00260] Example 4: Validation of NF-κΒ pathway members as mediators of resistance to MAPK pathway inhibitors

[00261] The identification of the NF-κΒ pathway through two independent, orthogonal approaches suggested to us that this pathway may be capable of modulating the sensitivity of melanomas to MAPK pathway inhibitors. To functionally validate this finding, we measured the effect of NF-κΒ activation on the half-maximal growth inhibitory concentrations (GI50) of PLX4720, AZD-6244, and Vertex 1 le, a selective ERK inhibitor³⁵, in four B-RAF^vmE mutant melanoma cell lines (A375, Colo679, UACC-62, and Sk-Mel-28). Overexpression of IKBKB or TRAF2 (Fig. 4a) conferred 1.5- to 10-fold GI50 shifts relative to MEK1 overexpression (which served as a neutral control; Supplementary Fig. 12). Drug resistance could also be induced in all cell lines by the addition of soluble TNFa, an established NF-KB agonist³⁶. In this assay system, the magnitude of the PLX4720 GI50 shift conferred by IKBKB or TRAF2 overexpression or TNFa addition was comparable to that observed with C- RAF overexpression, a well-established mediator of resistance to RAF inhibitors

(Supplementary Fig. 12)³³. Immunoblot analysis indicated that overexpression οΐΙΚΒΚΒ or TRAF2 or exogenous TNFa were sufficient to activate RelA phosphorylation (a commonly utilized measure of NF-κΒ pathway activity³⁶), but did not fully reactivate ERK phosphorylation in the presence of PLX4720 (Supplementary Fig. 13). NF-κΒ stimulation was also associated with resistance to PLX4720-induced apoptosis but not to bypass of drug- induced cell cycle arrest (Supplementary Fig. 14).

[00262] Our findings raised the possibility that NF-κΒ activity might predict clinical responses to MAPK inhibitors. As an initial test of this hypothesis, we used human-derived B-RAF^V600E malignant melanoma short-term cultures. We first queried steady state gene expression data from 29 cultures for signatures of NF-κΒ pathway activity³⁷. We selected four cultures that we predicted, on the basis of high NF-κΒ activity, to be resistant to MAPK inhibitors and four that we predicted, on the basis of low NF-κΒ activity, to be sensitive to MAPK inhibitors. Strikingly, all four high NF-κΒ cultures were strongly resistant to MAPK pathway inhibitors, while all four cultures with lowNF-κΒ activity were sensitive to MAPK inhibitors as expected (Fig. 4b; matching score = 0.906, p-value = 0.0086). Together, these data provide evidence that NF-κΒ pathway activity may predict clinical efficacy of MAPK inhibitors and play a functional role in the responses of patients to these drugs.

[00263] DISCUSSION

[00264] Our results demonstrate the scalability and screening applications of the MicroSCALE platform. To the best of our knowledge, MicroSCALE is the first miniaturized screening technology that combines efficient, selectable transgene expression, spatially confined infection and cell adhesion, scalable production, and rapid quantitative analysis, rendering it compatible with a broad range of adherent mammalian cell lines and assay formats. As such, we anticipate that this technology may broaden the scope of functional genomic screens, particularly those that require large combinations of cell lines, perturbations, and assay outputs or those involving cell line-, reagent-, or resource-limited settings.

[00265] Drug modifier screens are one attractive application of MicroSCALE technology because they have the potential to systematically reveal the genes and pathways that modulate drug sensitivity¹⁵'^42"43. To date, large-scale genetic modifier screens in mammalian cells have been limited in part by the practical limitations associated with screening across many drugs or cell lines using existing screening technologies. Using MicroSCALE to overcome this limitation, we generated kinome-wide drug modifier profiles across multiple classes of inhibitors, revealing several new insights. First, modifier screens reliably uncovered genes and pathways whose activatio confers drug resistance in a target- and pathway-selective manner, including some which we know to be clinically relevant (ex: C- RAF, SRC kinases, and C07)'⁵'³³'⁴⁴. Second, without wishing to be bound by any theory, our evidence suggests that drugs targeting common biochemical nodes or pathways will exhibit highly similar modifier profiles, reflecting their shared mechanisms of action, while those targeting distinct pathways will exhibit unique profiles, reflecting distinct mechanisms of action. This finding suggests that, when performed in sufficient scale (e.g., across many compounds) ORF-based modifier screens may ultimately be used to functionally annotate heretofore uncharacterized small molecule probes emerging from drug discovery pipelines^{36" '}". Third, our observation that inhibitors of the MAP and m^'fOR/PI3K signaling pathways have almost entirely non-overlapping modifier profiles suggests that the spectrum of events which can confer cross-resistance to inhibitors of independent signaling pathways may be considerably narrower than the spectrum of events which can confer cross-resistance to multiple compounds targeting the same pathway. This observation provides empirical support for the idea of testing combination therapies that inhibit independent signaling pathways as a means of preventing the emergence of drug resistance.

|00266| By integrating our large-scale functional screening data with steady-state gene expression and pharmacological sensitivity profiling, we nominated NF-κΒ pathway components as mediators of resistance to MAPK pathway inhibitors. These data are consistent with the finding that TNFa can block apoptosis induced by MEK1/2 inhibitors⁴⁵, as well as a second recent report demonstrating that NF-κΒ may mediate resistance to EGFR inhibitors in EGFR-mutent lung cancers⁴⁶. These findings may inform upcoming clinical trials for melanoma and other cancers that utilize targeted and immunotherapy drug combinations which impinge on both MAPK and NF- κΒ signaling^47"50. Overall, deployment of the approaches described here holds considerable potential for the scalable interrogation of many phenotypes linked to human biology and disease.

[00267] METHODS

[00268] Cell lines and reagents. A375, Colo679, UACC-62, Malme-3M, WM-793, WM- 1716, WM-1745, WM-1852, WM-1930, Sk-Br-3, HCC-827, UACC-812, ZR-75-1 , WM- 3627, WM-451Lu, WM-1862, and WM-3163 were grown in RPMI with 10% FBS and 1% penicillin/streptomycin. Sk-Mel-28, Sk-Mel-5, Lox IMVI, IGR-39, Hs294T, A2058, A549, U2-OS, U87, SW620, HBL-100, MCF-7, PC3, MDA-MB-231 , MDA-MB-453, SW480, FlCT-1 16, DDLS8817, LPS141 , and p53^"'^" mouse embryonic fibroblasts were grown in DMEM with 10% FBS and 1% penicillin/streptomycin. HeLa, 786-0, HepG2, DU145, 90- 8T, 293T, and Panc- 1 were grown in DMEM with 10% heat inactivated FBS and 1% penicillin/streptomycin. WM-1 15 and RPMI-7951 were grown in MEM with 10% FBS and 1% penicillin/streptomycin. Human mammary epithelial cell derivatives were grown in serum-free MEGM media as previously described⁵¹. PLX4720, GDC-0879, AZD-6244, PD- 0325901, and BEZ-235 were purchased from Selleck Chemicals. Torinl was obtained from N.S. Gray (Dana-Farber Cancer Institute). Compound 1 le was a gift from Vertex Pharmaceuticals.

[00269] Immunoblots and immunofluorescence. Immunoblotting was performed as previously described^{1 5} and blots were probed with primary antibodies recognizing phospho- RelA (Ser536, 1 : 1 ,000, Cell Signaling), RelA (1 : 1 ,000, Cell Signaling), phospho-ERKl/2 (Thr202/Tyr204, 1 : 1 ,000, Cell Signaling), and ERK1 /2 (1 : 1 ,000, Cell Signaling).

Immunofluorescence was performed by fixing slides with 3.7% paraformaldehyde, permeabilizing with 0.1 % Triton X-100, and staining at indicated dilutions in 1% bovine serum albumin. Immunofluorescence stains were Hoechst (1 : 10,000, Invitrogen), Syto82 (1 :5,000, Invitrogen), Alexa 546-Phalloidin (1 : 1 ,000, Invitrogen), phospho-S6 (Ser235/236, 1 : 1 ,000, Cell Signaling), and p24 (1 :500, ZeptoMetrix).

[00270] Image acquisition and analysis. Images in Figures 1 , 2(a-c), and Supplementary Figures 1 , 2, 4, and 6 were acquired using an Axiovert 200 microscope (Carl Zeiss) and analyzed using NIH Image J software. Images of Syto82-stained arrays and associated data in Figures 2(d-f), 3, and Supplementary Figure 5 were obtained using an Axon GenePix 4000B microarray scanner and spot intensities were analyzed using GenePix analysis software.

[00271] Production of MicroSCALE slides. Lentiviruses expressing shRNAs or full- length ORFs were produced and titered in high-throughput 96-well format as previously described³'⁶'¹⁵. Stock solutions of polybrene (Sigma) and chondroitin sulfate (Sigma) were dissolved at 8 mg/mL in PBS and sterile filtered. The two solutions were sequentially added to 1 mL of each lentiviral supernatant (in deep, v-bottom 96 well plates) to yield a final concentration of each polymer of 400 μ /ηιΙ_^. Solutions were then incubated for 15 min at room temperature. Plates were next centrifuged at 1 150g for 20 min, after which supernatants were aspirated using a multichannel wand aspirator. Thirty microliters of lentivirus printing buffer (containing 0.4M HEPES, 1.23M KC1, 12.5 mg/mL trehalose, and 12 μί¾/ηι!_^ protamine sulfate, pH adjusted to 7.3) was added to each well and lentiviral pellets were mechanically resuspended and transferred to 384-well, round-bottom source plates for printing. All fluid handling was performed using a Janus automated liquid handling workstation. MicroSCALE slides were printed using a two stage process onto polyacrylamide hydrogel-coated glass slides (CodeLink®, Surmodics) with an Aushon 2470 microarray printer. First, gelatin (Type B, bovine, Sigma) was dissolved at 2 mg/mL in deionized water containing 0.1% glycerol and printed to yield cell adhesive islands, then individual, concentrated lentivirus preparations were printed directly on top of gelatin features. During the technological development phase of this work, solid pins of multiple sizes were used, resulting in features sizes ranging from 300-600 μηι in diameter (spot sizes are indicated in each figure). Kinome ORF MicroSCALEs were printed using a pin that yields features that are approximately 600 μπι in diameter with 750 μιη center-center spacing. After printing, slides were stored in vacuum-sealed bags at -80° C until ready to use.

[00272] MicroSCALE functional assays and high-throughput screens. Slides were thawed at room temperature, blocked for 30 min using DMEM + 10% FBS, and seeded with 1-5 x 10⁵ cells per slide in 4-well slide chambers. Cells were allowed to attach and become infected on MicroSCALE features overnight and were then selected with puromycin (2.5 μg/mL) or blasticidin (10 g/mL). For assays described in Figure 1, Figure 2(a-c), Supplementary Figures 1,2, 4, and 6, and Supplementary Table 2, cells were incubated, fixed, stained, and imaged as indicated. For high-throughput ORF modifier screens, cells were selected for 2 d with blasticidin, then incubated for 5-7 d in normal growth medium containing blasticidin and the indicated drugs (PLX4720 (1 μΜ), GDC-0879 (1 μΜ), AZD- 6244 (250 nM), PD-0325901 (250 nM), BEZ-235 (200 nM), Torinl (200 nM), or vehicle (DMSO)). Slides were then stained with Syto82 and imaged as described above.

[00273] Analysis of screening data. Raw values corresponding to Syto82 fluorescence intensity on each MicroSCALE feature were first normalized by applying a local median smoothening algorithm which calculated the difference between the raw intensity of each feature and the median intensity of its six nearest neighbors in each X-Y direction. This approach, commonly used in the analysis of DNA or protein microarrays, adjusts for variations in cell density or autofluorescence across the slide surface⁵². Individual feature values were then normalized to the median value on the slide, then to the average normalized value of the same feature on 4 replicate, vehicle-treated slides (Viability Score). Finally individual spot viability scores were Z-transformed to indicate their distance (in number of standard deviations) from the array mean and the average value for 2-3 replicate spots per ORF was calculated⁵³. Hierarchical clustering of these data was performed using Cluster 3.0 and visualized using Java Treeview.

[00274] Secondary growth inhibition assays. ORF-expressing lenti iruses used in secondary assays were produced as previously described¹'. Cells were infected at a 1:10- 1 :20 dilution of virus in 6- well plates in the presence of 7.5 μ^/πά polybrene and centrifuged at 1200g for 1 h at 37° C. Twenty-four hours after infection blasticidin (10 μ^ηιΐ) was added and cells were selected for 48 h. Cells were then trypsinized, counted, and seeded into 96- well plates at 2,000 cells/well for growth inhibition assays. Twenty-four later, DMSO or concentrated serial dilutions of indicated drugs (in DMSO) were added to cells (1 :1 ,000) to yield final drag concentrations of 100, 10, 1, 0.1, 0.01, 0.001, 0.0001 or 0.00001 μΜ. Cell viability was measured 4 d after drug addition using the Cell Titer Glo® luminescent viability assay (Promega). Viability was calculated as the percentage of control (untreated cells) after background subtraction. A minimum of six replicates were performed for each cell line/ORF/drug/concentration. Data from growth-inhibition assays were displayed using GraphPad Prism 5 for Windows (GraphPad). GI50 values were determined as the drug dose corresponding to half-maximal growth inhibition as previously described¹⁵. Growth curves that crossed the 50% inhibition point at or above 10 μΜ have G150 values annotated simply as >10 μΜ. GI50 values for unmodified parental cells were determined by seeding cells directly into 96-well plates and conducted assays as described above. For single-dose studies, cells were seeded at 500 cells/well in 96-well plates, infected at 24 h using a 1 :20 dilution of virus in the presence of 7.5 μ§ ιτιΙ. polybrene followed by centrifugation at 1200g for 1 h at 37° C, selected with 10 μg mL blasticidin for 48 h, then treated with drug or vehicle and assayed for viability using Cell Titer Glo® as above following a 4 d incubation.

Viability was calculated as the percentage of control (vehicle treated cells) after background subtraction.

[00275| Analysis of cell cycle and apoptosis. Cells were seeded into 10 cm dishes on day 0, treated as indicated with PLX4720 (1 μΜ), TNFa (25 ng/mL), or vehicle on day 1, and analyzed on day 3. For the analysis of cell cycle distributions, cells were fixed with 80% ethanol in H₂0 and stained with 50 μg mL propidium iodide (BD Pharmingen) containing 0.1 mg/mL RNAse A and 0.05% Triton X-100. For the analysis of Annexin V staining, cells were suspended in Annexin V binding buffer (10 mM HEPES, 140 mM NaCl, 2.5 mM CaCk, pH 7.4) containing Annexin V-APC (BD Pharmingen) and 50 μg/mL propidium iodide. For both analyses, a minimum of 50,000 events were counted per sample. Cell cycle data were analyzed using ModFit software. Annexin V staining was analyzed using FlowJo software, with Annexin V-positive cells defined as those exhibiting Annexin V staining intensities exceeding 99.9% of cells in a Pi-only control sample.

[00276] Cancer Cell Line Encyclopedia (CCLE) data. Gene expression and pharmacological sensitivity data for 25 BRAF^r600E melanoma cell lines were obtained by the Cancer Cell Line Encyclopedia (CCLE) project, a collaboration between the Broad Institute, the Novartis Institutes for Biomedical Research (NIBR) and the Genomics Institute of the Novartis Research Foundation (GNF). Gene expression data used for this study are available online at the Broad Institute CCLE (http://www.broadinstitute.org/ccle/home).

[00277] Gene expression analysis. The single-sample GSEA enrichment scores used in Figs. 3d and 4b were obtained as described in Barbie et al⁴. Briefly, for every gene- expression sample profile the values were first rank-normalized and sorted, and then a single- sample enrichment score for each gene set is computed based on the integrated difference between the empirical cumulative distribution functions of the genes in the gene set vs. the rest. This procedure is similar to the computation of standard Gene Set Enrichment Analysis³⁴ but is based on absolute rather than differential expression. Published details of this method and other applications are available^34,54"55. Matching scores between PLX4720 vs. single- sample GSEA gene set scores (Fig. 3d), were obtained by a normalized and rescaled mutual information estimate. Briefly, we considered the differential mutual information⁵⁶ between two continuous vectors x (target, e.g. PLX4720 resistance) and y (feature, e.g. a NF-κΒ gene set):

and estimated this quantity using a kernel- density estimate of the joint distribution P(x.y). The discrete data was smoothed via a Gaussian kernel, with width determined by a cross-validation bandwidth estimation⁵⁷ at each data point (*¾ >'£>, and Pfr. ) was found by summing overall densities over a discrete grid information, /(^■*^"> '), was then

matching score was obtained by rescaling v) using the normalized mutual information of * (the target) with itself, and adding a "direction" (+/-) _{s(x m} sign[p(x,y))U(x, y)

according to the Pearson correlation pbe. y) ^' '^"' U(x, x) . A perfect match

(antimatch) corresponds to a score of +1 (-1) and a random match to 0. The significance of a given match SO . > was estimated by a permutation test where the values of * are randomly permutated 10,000 times, and a nominal p- value was computed according to how many times the matching scores of the random permutations are higher than the actual score. This matching score S(x, >') has advantages over other metrics including increased sensitivity to non-linear associations and wider dynamic range at the top of the matching scale.

[00278] Statistics. Results are expressed as the mean ± standard deviation. For comparisons between two groups, P values were calculated using unpaired, two-tailed

Student's t-tests.

[00279] Supplementary Table 3: Constructs present on Kinome ORF MicroSCALEs

(Johanneswn ot

lus ttw additional

] REFERENCES

Shapira, S.D., et al A physical and regulatory map of host-influenza interactions reveals pathways in H1N1 infection. Cell 139, 1255-1267 (2009).

Brass, A.L., et al. Identification of host proteins required for HIV infection through a functional genomic screen. Science 319, 921 -926 (2008).

Moffat, J., et al. A lentiviral RNAi library for human and mouse genes applied to an arrayed viral high-content screen. Cell 124, 1283-1298 (2006).

Barbie, D.A., et al. Systematic RNA interference reveals that oncogenic KRAS- driven cancers require TBK1 , Nature 462, 108-1 12 (2009).

Mohr, S., Bakal, C. & Perrimon, N. Genomic Screening with RNAi: Results and Challenges, in Ann Rev Biochem, Vol. 79 37-64 (Annual Reviews, Palo Alto, 2010). Root, ! >.!· .. Hacohen, N., Hahn, W.C., Lander, E.S. & Sabatini, D.M. Genome-scale loss-of-function screening with a lentiviral RNAi library. Nat Methods 3, 715-719 (2006).

Schlabach, M.R., et al. Cancer proliferation gene discovery through functional genomics. Science 319, 620-624 (2008).

Silva, J.M., et al. Profiling essential genes in human mammary cells by multiplex RNAi screening. Science 319, 617-620 (2008).

Luo, B„ et al. Highly parallel identification of essential genes in cancer cells. Proc. Natl. Acad. Sci. U.S.A. 105, 20380-20385 (2008).

Berns, K., et al. A large-scale RNAi screen in human cells identifies new components of the p53 pathway. Nature 428, 431-437 (2004).

Ziauddin, J. & Sabatini, D.M. Microarrays of cells expressing defined cDNAs. Nature 411, 107-110 (2001).

Flaim, C.J., Chien, S. & Bhatia, S.N. An extracellular matrix microarray for probing cellular differentiation. Nat Methods 2, 1 19-125 (2005).

Reiser, J. Production and concentration of pseudotyped HIV- 1 -based gene transfer vectors. Gene Ther 7, 910-913 (2000).

Le Doux, J.M., Landazuri, N., Yarmush, M.L. & Morgan, J.R. Complexation of retrovirus with cationic and anionic polymers increases the efficiency of gene transfer. Hum Gene Ther 12, 161 1-1621 (2001).

Johannessen, CM., et al. COT drives resistance to RAF inhibition through MAP kinase pathway reactivation. Nature 468, 968-972 (2010).

Lindquist, R.A., et al. Genome-scale RNAi on living-cell microarrays identifies novel regulators of Drosophila melanogaster TORC1-S6K pathway signaling. Genome Res 21, 433-446 (2011).

Engelman, J.A., et al. MET amplification leads to gefitinib resistance in lung cancer by activating ERBB3 signaling. Science 316, 1039-1043 (2007).

Kim, D.H., et al. niTOR interacts with raptor to form a nutrient-sensitive complex that signals to the cell growth machinery. Cell 110, 163-175 (2002).

Karaman, M.W., et al. A quantitative analysis of kinase inhibitor selectivity. Nat. Biotechnol. 26, 127-132 (2008).

Tsai, J., et al. Discovery of a selective inhibitor of oncogenic B-Raf kinase with potent antimelanoma activity. Proc. Nail. Acad. Sci. U.S.A. 105, 3041 -3046 (2008). Bollag, G„ et al. Clinical efficacy of a RAF inhibitor needs broad target blockade in BRAF-mutant melanoma. Nature 467, 596-599 (2010). Flaherty, K.T., et al. Inhibition of mutated, activated BRAF in metastatic melanoma, N. Engl. J. Med. 363, 809-819 (201 0).

Emery, CM., et al. MEK1 mutations confer resistance to MEK and B-RAF inhibition. Proc. Natl. Acad. Set. U.S.A. 106, 2041 1-2041 6 (2009).

Lee, J.W., el al. Somatic mutations of F.RBB2 kinase domain in gastric, colorectal, and breast carcinomas. Clin. Cancer Res. 12, 57-61 (2006).

TCGA_Research_Network. Comprehensive genomic characterization defines human glioblastoma genes and core pathways. Nature 455, 1061-1068 (2008).

Nazarian, R., et al. Melanomas acquire resistance to B-RAF(V600E) inhibition by RTK or N-RAS upregulation. Nature 468, 973-977 (2010).

Villanueva, J., et al. Acquired Resistance to BRAF Inhibitors Mediated by a RAF Kinase Switch in Melanoma Can Be Overcome by Cotargeting MEK and IGF- 1 R/PI3K. Cancer Cell 18, 683-695 (201 0).

Hoeflich, K.P., et al. Antitumor efficacy of the novel RAF inhibitor GDC-0879 is predicted by BRAFV600E mutational status and sustained extracellular signal- regulated kinase/mitogen-activated protein kinase pathway suppression. Cancer Res. 69, 3042-3051 (2009).

Yeh, T.C., et al. Biological characterization of ARRY-142886 (AZD6244), a potent, highly selective mitogen-activated protein kinase kinase 1 /2 inhibitor, Clin Cancer Res 13, 1576-1583 (2007).

Thompson, N. & Lyons, J. Recent progress in targeting the Raf/MEK/ERK pathway with inhibitors in cancer drag discovery. Curr Opin Pharmacol 5, 350-356 (2005). Thoreen, C.C., et al. An ATP-competitive mammalian target of rapamycin inhibitor reveals rapamycin-resistant functions of mTORC l . ./ Biol Chem 284, 8023-8032 (2009).

Maira, S.M., et al. Identification and characterization of NVP-BEZ235, a new orally available dual phosphatidyliiiositol 3-kinase/manimalian target of rapamycin inhibitor with potent in vivo antitumor activity, Mol Cancer Ther 7, 1851 - 1863 (2008).

Montagut, C, et al. Elevated CRAF as a potential mechanism of acquired resistance to BRAF inhibition in melanoma. Cancer Res. 68, 4853-4861 (2008).

Subramanian, A„ et al. Gene set enrichment analysis: A knowledge-based approach for interpreting genome-wide expression profiles. Proc. Natl. Acad. Sci. U.S.A. 102, 15545-15550 (2005).

Aronov, A.M., et al. Structure-Guided Design of Potent and Selective Pyrimidylpyrrole Inhibitors of Extracellular Signal-Regulated Kinase (ERK) Using Conformational Control. ./ Med. Chem. 52, 6362-6368 (2009).

Karin, M. & Ben-Neriah, Y. Phosphorylation meets ubiquitination: The control of NF-kappa B activity. Anna. Rev. Immunol. 18, 621 -+ (2000).

Lin, W.M., et al. Modeling genomic diversity and tumor dependency in malignant melanoma. Cancer Res 68, 664-673 (2008).

Neumann, B„ et al. High-throughput RNAi screening by time-lapse imaging of live human cells. Nat Methods 3, 385-390 (2006).

Silva, J.M., Mizuno, H., Brady, A., Lucito, R. & Harmon, G.J. RNA interference microarrays: High-throughput loss-of-function genetics in mammalian cells. Proc. Natl Acad. Sci. U.S.A. 101, 6548-6552 (2004).

Mousses, S„ et al. RNAi microarray analysis in cultured mammalian cells. Genome Res. 13, 2341-2347 (2003). Bailey, S.N., Ali, S. ., Carpenter, A.E., Higgins, CO. & Sabatini, D.M. Micro-arrays of ientivi uses for gene function screens in immortalized and primary cells. Nat Methods 3, 1 17-122 (2006).

Lokey, R.S. Forward chemical genetics: progress and obstacles on the path to a new pharmacopoeia. Curr Opin Chem Biol 7, 91 -96 (2003).

Parsons, A.B., et al. Exploring the mode-of-action of bioactive compounds by chemical-genetic profiling in yeast. Cell 126, 611-625 (2006).

Zhang, S., et al. Combating trastuzurnab resistance by targeting SRC, a common node downstream of multiple resistance pathways. Nat Med 17, 461-469 (201 1).

Gray-Schopfer, V.C., f arasarides, M., Hayward, R. & Marais, R. Tumor necrosis factor-alpha blocks apoptosis in melanoma cells when BRAF signaling is inhibited. Cancer Res. 67, 122-129 (2007).

Bivona, T.G., et al. FAS and NF-kappa B signalling modulate dependence of lung cancers on mutant EGFR. Nature 471. 523-526 (201 1).

Perrin, P.J., Maldonado, J.H., Davis. T.A., June, C.H. & Racke, M. . CTLA-4 blockade enhances clinical disease and cytokine production during experimental allergic encephalomyelitis. J. Immunol. 157, 1333-1336 (1996).

Yuan, J.D., et al. CTLA-4 blockade enhances polyfunctional NY-ESO-1 specific T cell responses in metastatic melanoma patients with clinical benefit. Proc. Natl. Acad. Sci. U.S.A. 105, 20410-20415 (2008).

Economou, J.S., et al. Production of tumor necrosis factor alpha and interferon gamma in interleukin 2-treated melanoma patients - Correlation with clinical toxicity. Cancer Immunol Immunother. 34, 49-52 (1991).

Boni, A., et al. Selective BRAF(V600E) Inhibition Enhances T-Cell Recognition of Melanoma without Affecting Lymphocyte Function. Cancer Res. 70, 5213-5219 (2010).

Chaffer, C.L., et al. Normal and neoplastic nonstem cells can spontaneously convert to a stem-like state. Proc. Natl. Acad. Sci. U.S.A. 108, 7950-7955 (201 1).

Irizarry, R.A., et al. Exploration, normalization, and summaries of high density oligonucleotide array probe level data. Bio statistics 4, 249-264 (2003).

Birmingham, A., et al. Statistical methods for analysis of high-throughput RNA interference screens. Nat Methods 6, 569-575 (2009).

Jagani, Z., et al. Loss of the tumor suppressor Snf5 leads to aberrant activation of the Hedgehog-Gli pathway. Nat Med 16, 1429-U1 112 (2010).

Tamayo, P., et al. Predicting Relapse in Patients With Medulloblastoma by Integrating Evidence From Clinical and Genomic Features. J. Clin. Oncol. 29, 1415- 1423 (201 1 ).

Cover, T. & Thomas, J. Elements of Information Theory, (John Wiley & Sons, Hoboken, NJ, USA, 2006).

Scott, D.W. Multivariate Density Estimation: Theory, Practice, and Visualization (John Wiley & Sons, Hoboken, NJ, USA, 1992).

Ming, L., Xin, C, Xin, L., Bin, M. & Vitanyi, P.M.B. The similarity metric. Information Theory. IEEE Transactions on 5(1, 3250-3264 (2004).

Kraskov, A., Stogbauer, H„ Andrzejak, R.G. & Grassberger, P. Hierarchical clustering using mutual information. Europhys. Lett. 70, 278-284 (2005). [00281] Those skilled in the art will recognize, or be able to ascertain using no more than routine experimentation, many equivalents to the specific embodiments of the invention described herein. The scope of the present invention is not intended to be limited to the Description or the details set forth therein. Articles such as "a", "an" and "the" may mean one or more than one unless indicated to the contrary or otherwise evident from the context. Claims or descriptions that include "or" between one or more members of a group are considered satisfied if one, more than one, or all of the group members are present in, employed in, or otherwise relevant to a given product or process unless indicated to the contrary or otherwise evident from the context. The invention includes embodiments in which exactly one member of the group is present in, employed in, or otherwise relevant to a given product or process. The invention also includes embodiments in which more than one, or all of the group members are present in, employed in, or otherwise relevant to a given product or process. Furthermore, it is to be understood that the invention encompasses all variations, combinations, and permutations in which one or more limitations, elements, clauses, descriptive terms, etc., from one or more of the claims (whether original or subsequently added claims) is introduced into another claim (whether original or subsequently added). In particular, any claim that is dependent on another claim can be modified to include one or more elements or limitations found in any other claim that is dependent on the same base claim. Furthermore, where the claims recite a product (e.g., an array) or composition, the invention provides methods of making the product composition, e.g., according to methods disclosed herein, and methods of using the product composition, e.g., for puiposes disclosed herein. Also, where the claims recite a method of making a product or composition, the invention provides products or compositions made according to the inventive methods and methods of using the composition, unless otherwise indicated or unless one of ordinary skill in the art would recognize that a contradiction or inconsistency would arise. It is contemplated that any one or more embodiments or aspects of the invention can be freely combined with one or more other embodiments or aspects whenever appropriate.

100282] Where elements are presented as lists, e.g., in Markush group format, each subgroup of the elements is also disclosed, and any element(s) can be removed from the group. For purposes of conciseness only some of these embodiments have been specifically recited herein, but the invention includes all such embodiments. It should also be understood that, in general, where the invention, or aspects of the invention, is/are referred to as comprising particular elements, features, etc., certain embodiments of the invention or aspects of the invention consist, or consist essentially of, such elements, features, etc,

[00283] Where numerical ranges are mentioned herein, the invention includes embodiments in which the endpoints are included, embodiments in which both endpoints are excluded, and embodiments in which one endpoint is included and the other is excluded. It should be assumed that both endpoints are included unless indicated otherwise. Furthermore, unless otherwise indicated or otherwise evident from the context and understanding of one of ordinary skill in the art, values that are expressed as ranges can assume any specific value or subrange within the stated ranges in different embodiments of the invention, to the tenth of the unit of the lower limit of the range, unless the context clearly dictates otherwise. Where phrases such as "less than X", "greater than X", or "at least X" is used (where X is a number or percentage), it should be understood that any reasonable value can be selected as the lower or upper limit of the range. It is also understood that where a list of numerical values is stated herein (whether or not prefaced by "at least"), the invention includes embodiments that relate to any intervening value or range defined by any two values in the list, and that the lowest value may be taken as a minimum and the greatest value may be taken as a maximum. Furthermore, where a list of numbers, e.g., percentages, is prefaced by "at least", the term applies to each number in the list. For any embodiment of the invention in which a numerical value is prefaced by "about" or "approximately", the invention includes an embodiment in which the exact value is recited. For any embodiment of the invention in which a numerical value is not prefaced by "about" or "approximately", the invention includes an embodiment in which the value is prefaced by "about" or "approximately". "Approximately" or "about" generally includes numbers that fall within a range of 1% or in some embodiments 5% or in some embodiments 10% of a number in either direction (greater than or less than the number) unless otherwise stated or otherwise evident from the context (e.g., where such number would impermissibly exceed 100% of a possible value). Any particular embodiment(s), aspect(s), or element(s) of the present invention may be explicitly excluded from any one or more of the claims.

Claims

We claim:

1. An array comprising a surface having multiple discrete features, wherein each feature comprises a ceil adhesive material and an agent, and wherein the features are separated by regions that inhibit mammalian cell migration.

2. The array of claim, wherein the agents are reversibly associated with the cell adhesive material.

3. The array of claim 1 , wherein the regions that inhibit mammalian cell migration

inhibit eukaryotic cell adhesion.

4. The array of claim 1 , wherein the regions that inhibit mammalian cell migration

comprise a cell adhesion resistant material.

5. The array of claim 1 , wherein the agents comprise nucleic acids, polypeptides, or small molecules.

6. The array of claim 1 , wherein the agents comprise viruses.

7. The array of claim 1 , wherein different features contain distinct agents of known identity to be contacted with cells.

8. The array of claim 1, wherein different features contain distinct agents of known identity whose effect on cell phenotype is to be assessed.

9. The array of claim 1 , wherein the agents comprise viruses, and wherein viruses in different features contain distinct nucleic acids of known identity to be introduced into eukaryotic cells.

10. The array of claim 1 , wherein the features have uniform size and shape.

1 1. The array of claim 1 , wherein the features are arranged in a regular pattern.

12. The array of claim 1, wherein the array comprises at least 50 features per square centimeter (cm) over an area of at least 10 cm .

13. The array of claim 1 , wherein the array comprises at least 1 ,000 features.

14. The array of claim 1 , wherein the features have a diameter of about 300 μηι - 600 μηι.

15. The array of claim 1 , wherein the edges of adjacent features are separated by a

distance of about 100 μπι - 200 μη .

16. The array of claim 1 , wherein the array comprises at least 300 features comprising distinct agents to be contacted with cells.

17. The array of claim 1 , wherein the array comprises at least 300 features comprising viruses containing distinct nucleic acids.

18. The array of claim 1 , wherein the surface is a glass or plastic surface.

19. The array of claim 1 , wherein the surface is the surface of a slide.

20. The array of claim 1 , wherein the agents comprise enveloped viruses.

21. The array of claim 1 , wherein the agents comprise lentiviruses.

22. The array of claim 1 , wherein the agents comprise retroviruses.

23. The array of claim 1 , wherein the agents contain an expression cassette that encodes a selectable marker, wherein the selectable marker is optionally a drug resistance marker.

24. The array of claim 1 , wherein the features comprise viruses deposited in a

composition comprising at least 10⁸ infectious units/milliliter.

25. The array of claim 1, wherein the features comprise viruses and negatively and

positively charged polyelectrolytes.

26. The array of claim 1 , wherein the cell adhesive material comprises a protein.

27. The array of claim 1 , wherein the cell adhesive material comprises a gelatin.

28. The array of claim 1 , wherein the region that inhibits mammalian cell migration

comprises a hydrogel.

29. The array of claim 1 , wherein the region that inhibits mammalian cell migration comprises a hydrogel comprising polyacrylamide.

30. The array of claim 1 , wherein the surface is at least partly coated with a cell adhesion resistant material, and features comprising a cell adhesive material are located on top of the cell adhesion resistant material.

3 1. The array of claim 1 , wherein the features are printed using a printer with solid pins.

32. The array of claim 1 , wherein at least 80% of the agents comprise nucleic acid

sequences that correspond to mammalian genes.

33. The array of claim 1 , wherein at least 80% of the agents comprise nucleic acid

sequences that correspond to human genes.

34. The array of claim 1 , wherein at least 80% of the agents comprise a nucleic acid sequence that, when expressed in a eukaryotic cell containing a gene comprising a corresponding sequence, inhibit a function of said gene in the eukaryotic cell.

35. The array of claim 1 , wherein at least 80% of the agents comprise a nucleic acid sequence that encodes a short hairpin RNAs (shRNAs).

36. The array of claim 1 , wherein at least 80%o of the agents comprise a nucleic acid sequence that encodes a protein.

37. The array of claim 1 , wherein at least 80%o of the agents comprise nucleic acid

sequences that encode, or inhibit expression of, proteins that belong to a selected category.

38. The array of claim 37, wherein the selected category is: enzymes, kinases,

transcription factors, cell surface receptors, G proteins, signal transduction pathway proteins, oncogenes, or tumor suppressor genes.

39. The array of claim 1 , wherein the agents include at least one agent that encodes a mutant form of a protein of interest.

40. The array of claim 1 , wherein the agents include at least one agent that encodes a mutant form of a protein of interest, wherein expression of said mutant form by a cell contributes to drug resistance.

41. The array of claim 1 , wherein the array is capable of supporting attachment and

proliferation of diverse mammalian cell lines to yield discrete spatially-defined regions of cells whose shape and size corresponds to the shape and size of said features, and wherein the array is substantially devoid of cells between said regions.

42. The array of claim 1 , wherein the array can be stably stored at -80°C for at least 8 months with substantially no detectable loss of activity.

43. A collection of arrays of claim 1 , wherein the collection includes at least 10,000

features containing distinct nucleic acids.

44. The collection of claim 43, wherein the collection includes features containing nucleic acids that correspond in sequence to at least 80% of the genes of a mammalian genome.

45. The collection of claim 43, wherein the collection includes features containing nucleic acids that correspond in sequence to at least 80% of the genes of the human genome.

46. A composition comprising: (a) the array of any of claims 1 - 40; and (b) eukaryotic cells.

47. The composition of claim 46, wherein the cells are mammalian cells.

48. The composition of claim 46, wherein the cells are human cells.

49. The composition of claim 46, further comprising cell culture medium that contains a test substance.

50. A kit comprising: (a) the array or collection of any of claims 1 - 45; and (b)

instructions for using the array to prepare an array of eukaryotic cells, wherein the instructions optionally include information regarding the identity of the agent located in different features.

51. The kit of claim 50, further comprising at least one item selected from the group consisting of: a printing buffer, a cell selection reagent, and a cell detection reagent.

52. A composition comprising: (a) polyelectrolyte-virus complexes; (b) an aqueous

medium; and (c) a salt, wherein the composition is substantially free of serum.

53. The composition of claim 52, wherein the viruses are lentiviruses.

54. The composition of claim 52, further comprising a stabilizing agent.

55. The composition of claim 44, further comprising an additional polycation, wherein the additional polycation is optionally distinct from a polycation present in said polyelectrolyte-virus complexes.

56. The composition of claim 52, further comprising a cell adhesive material.

57. The composition of claim 52, further comprising a gelatin.

58. A collection comprising at least 50 compositions as set forth in claim 52, wherein the virus in each of at least 50 compositions comprise distinct genes to be introduced into eukaryotic cells, and

59. The collection of claim 58, wherein the compositions are in wells of a multiwell plate.

60. A method of making an array of eukaryotic cells comprising: (a) providing an array of any of claims 1 - 45; and (b) contacting the surface of the array with eukaryotic cells and a cell culture medium under conditions suitable for cell adhesion.

61. The method of claim 60, wherein the agents comprise viruses, and wherein the

method comprises maintaining the array under appropriate conditions so that viruses are introduced into the eukaryotic cells at the locations of said features.

62. The method of claim 60, wherein the eukaryotic cells are mammalian cells.

63. The method of claim 60, comprising maintaining the array for at least 24 hours in selective cell culture medium.

64. The method of claim 60, further comprising contacting the cells with a test substance.

65. The method of claim 60, further comprising assessing a phenotype of the cells located atop said features.

66. The method of claim 60, further comprising scanning the array with a microarray scanner, thereby gathering information pertaining to a phenotype of the cells located at said features.

67. An array comprising virus-infected eukaryotic cells, wherein the array is prepared according to the method of claim 60.

68. A method of making an array for infection or transfection of eukaryotic cells, the method comprising: (a) providing a surface comprising spots that comprise a cell adhesive material, wherein the spots are separated by a cell adhesion resistant material; and (b) depositing an agent on top of each spot.

69. The method of claim 68, wherein the surface is at least partly coated with the cell adhesion resistant material, and wherein the spots that comprise a cell adhesive material are located at multiple discrete locations on the coating of cell adhesion resistant material.

70. The method of claim 68, wherein step (a) comprises: (i) providing a surface that is at least partly coated with a cell adhesion resistant material; and (ii) depositing a cell adhesive material on top of the cell adhesion resistant material at multiple discrete locations so as to form spots comprising said cell adhesive material.

71. The method of claim 68, wherein the agents are viruses that comprise a nucleic acid to be introduced into eukaryotic cells.

72. The method of claim 68, wherein step (b) comprises depositing viruses on the spots in a composition comprising positively and negatively charged polyelectrolytes.

73. The method of claim 72, wherein the viruses are deposited in a liquid comprising at least 10 infectious units of virus per milliliter of the composition.

74. The method of claim 72, wherein the composition is substantially free of serum.

75. The method of claim 72, comprising preparing the viruses by a method comprising contacting the viruses with positively and negatively charged polyelectrolytes.

76. The method of claim 68, further comprising freezing the array.

77. The method of claim 68, further comprising plating mammalian cells atop the array.

78. An array prepared according to method of claim 68.

79. A method of preparing an array for infection or transfection of eukaryotic cells, the method comprising: (a) providing a surface at least partly coated with a cell adhesion resistant material; and (b) depositing a composition comprising a cell adhesive material and an agent at multiple discrete locations on said cell adhesion resistant material.

80. The method of claim 79, wherein the agents are viruses that comprise a nucleic acid to be introduced into eukaryotic cells.

81. The method of claim 79, wherein step (b) comprises depositing viruses on the spots in a composition comprising positively and negatively charged polyelectrolytes.

82. The method of claim 79, wherein the viruses are deposited in a liquid comprising at least l O⁸ infectious units of virus per milliliter of the composition.

83. The method of claim 79, wherein the composition is substantially free of serum.

84. The method of claim 79, comprising preparing the viruses by a method comprising contacting the viruses with positively and negatively charged polyelectrolytes.

85. The method of claim 79, further comprising freezing the array.

86. The method of claim 79, further comprising plating mammalian cells atop the array.

87. An array prepared according to method of claim 79.

88. A method of making an array, the method comprising steps of: (a) forming multiple cell adhesive regions atop a cell adhesion resistant surface; and (b) depositing an agent on each of said cell adhesive regions.

89. The method of claim 88, wherein the cell adhesion resistant surface comprises polylacrylamide.

90. The method of claim 88, wherein the cell adhesive regions comprise gelatin.

91. The method of claim 88, wherein step (b) comprises depositing lentiviruses in a

polyectrolyte complex.

92. The method of claim 88, further comprising plating mammalian cells atop the array.

93. An array comprising virus-infected eukaryotic cells, the array comprising; a surface having multiple discrete features comprising a cell adhesive material, wherein the features are separated by a cell adhesion resistant material, and wherein virus-infected eukaryotic cells are growing atop said features in spatially defined regions corresponding in shape and size to said features and separated by regions that are substantially devoid of cells.

94. The array of claim 93, wherein the surface is at least partly coated with the cell

adhesion resistant material, and the cell adhesive material is located on top of the cell adhesion resistant material.

95. The array of claim 93, wherein the eukaryotic cells are mammalian cells.

96. The array of claim 93, wherein the eukaryotic cells are human cells.

97. The array of claim 93, wherein the viruses are lentiviruses.

98. The array of claim 93, wherein virus- infected eukaryotic ceils are proliferating atop least 80% of said features.

99. The array of claim 93, wherein vims-infected eukaryotic cells arc proliferating atop least 90% of said features.

100. The array of claim 93, wherein at least 90% of control features lacking viruses are substantially devoid of cells.

101. A method comprising: (a) providing an array of claim 93, wherein expression of the nucleic acid by the virus-infected eukaryotic cells in a spatially defined region increases or decreases expression or activity of a gene product in said cells; (b) contacting the array with a test compound; and (c) assessing at least one phenotype of cells located in said spatially defined regions.

102. The method of claim 101 , further comprising: (d) identifying one or more spatially defined regions in which at least one phenotype is altered in said cells as compared with control cells, thereby identifying a gene product that modulates the response of eukaryotic cells to the test compound.

103. The method of claim 101 , wherein assessing the phenotype comprises measuring expression of a reporter in the spatially defined regions.

104. The method of claim 101 , wherein assessing the phenotype comprises measuring fluorescence intensity associated with the spatially defined regions.

105. The method of claim 101 , wherein the phenotype is assessed using a microarray

scanner.

106. The method of claim 101 , wherein the phenotype is indicative of cell growth or

proliferation.

107. The method of claim 101 , wherein the method identifies a gene product that

modulates the sensitivity of eukaryotic cells to the test agent.

108. The method of claim 1 01 , wherein at least 80% of the viruses contain nucleic acids comprising a sequence that encodes a eukaryotic protein.

109. The method of claim 1 01 , wherein at least 80% of the viruses contain nucleic acids that, when expressed in a eukaryotic cell that expresses a gene comprising a corresponding sequence, inhibit a function of said gene in the eukaryotic cell.

1 10. The method of claim 101 , wherein the test compound is a small molecule.

1 1 1. The method of claim 101 , wherein the eukaryotic cells are mammalian cells.

1 12. The method of claim 1 01 , wherein the test compound is an approved therapeutic agent or candidate therapeutic agent.

1 1 3. The method of claim 1 01 , wherein steps (a)-(d) are repeated using multiple different concentrations of the test agent.

1 14. The method of claim 101 , wherein steps (a)-(d) are repeated using multiple different test compounds that inhibit the same protein.

1 15. The method of claim 1 01 , wherein steps (a)-(d) are repeated using multiple different test compounds that inhibit proteins in the same biological pathway.

1 16. The method o claim 1 01 , wherein the method comprises receiving the test compound or information identifying the test compound from a requestor; and reporting at least some results of the phenotypic assessment to the requestor.

1 1 7. A method of inhibiting resistance of a tumor or tumor cell to a MAP pathway

inhibitor, the method comprising contacting the tumor or tumor cell with a MAPK pathway inhibitor and an NF-κΒ pathway inhibitor, thereby inhibiting resistance of the tumor cell to the MAPK pathway inhibitor.

1 1 8. The method of claim 1 17, wherein the tumor or tumor cell is a melanoma or

melanoma cell, respectively.

1 19. A method of treating a subject in need of treatment for a tumor, the method

comprising administering (i) a MAPK pathway inhibitor; and (ii) an NF-κΒ pathway inhibitor, Src pathway inhibitor, mTOR pathway inhibitor, pI3 kinase pathway inhibitor, or protein kinase C (PKC) pathway inhibitor to the subject.

1 20. The method of claim 1 1 9, wherein the method comprises treating the subject with (i) a MAPK pathway inhibitor; and (ii) an NF-κΒ pathway inhibitor.

121 . The method of claim 1 1 9, wherein the MAPK pathway inhibitor inhibits Raf kinase.

122. The method of claim 1 1 9, wherein the NF-κΒ pathway inhibitor, Src pathway

inhibitor, mTOR pathway inhibitor, pI3 kinase pathway inhibitor, or protein kinase C (PKC) pathway inhibitor binds to and inhibits a protein kinase component of the NF- KB pathway, Src pathway, mTOR pathway, pI3 kinase pathway, or protein kinase C (PKC) pathway.

123. The method of claim 1 19, wherein the tumor is a melanoma.

124. A method of assessing the likelihood that a tumor or tumor cell is resistant to MAPK pathway inliibition, the method comprising the step of measuring the activity of the NF-κΒ pathway in the tumor or tumor cell, wherein increased activity of the NF~KB pathway as compared with a suitable control indicates an increased likelihood that the tumor or tumor cell is resistant to MAPK pathway inhibition.

125. The method of claim 124, wherein the tumor or tumor cell is a melanoma or

melanoma cell, respectively.