RU2805952C2 - Phasing correction - Google Patents

Abstract

FIELD: biotechnology; medicine.

SUBSTANCE: invention is a method of determining corrected color values from image data obtained during a base sensing cycle using a nucleic acid sequencer comprising an image acquisition system, one or more processors, and a storage device, comprising the following: (a) obtaining an image of a substrate containing a plurality of base sensing sites; nucleic acids, the portions having colors representing types of nucleic bases; (b) measuring color values of a plurality of regions from said substrate image; (c) storing the color values in a processor buffer of one or more sequencer processors; (d) retrieving phasing-corrected color values of a plurality of regions where said phasing-corrected color values were stored in the sequencer memory during the immediately preceding base recognition cycle; and (e) determining the corrected color values from the color values in the processor buffer and the phasing corrected color values stored from the immediately preceding cycle.

EFFECT: invention makes it possible to determine corrected color values from image data obtained using a nucleic acid sequencer during a base recognition cycle.

18 cl, 11 dwg

Description

Cross references to related applications

[1] This application claims benefit from U.S. Provisional Patent Application No. 62/443,294, filed January 6, 2017, entitled “PHASING CORRECTION,” which is incorporated herein by reference in its entirety and for all purposes.

BACKGROUND OF THE ART

[2] The present disclosure relates to nucleic acid sequencing. More specifically, the present disclosure relates to systems and methods for real-time sequencing with phase correction.

[3] Multiple copies of a nucleic acid molecule having the same sequence (possibly with limited variations unintentionally introduced during sample processing) are analyzed together at a specific location in a flow cell or other support. Sufficient copies are used to generate sufficient signal to ensure reliable base recognition. A set of nucleic acid molecules in a site is called a cluster.

[4] Phasing is an off-target artifact that results from sequencing multiple nucleic acid molecules in a cluster. Phasing is the degree to which signals, such as fluorescence from individual molecules in a cluster, lose synchronization with each other. Often the term “phasing” is used to refer to interference from some molecules that are lagging, and the term “pre-phasing” (advanced) is used to refer to interference from other molecules that are leading. Together, phasing and phasing forward determine how well the sequencing device (sequencer) performs and the chemical analysis is performed.

BRIEF DESCRIPTION OF THE INVENTION

[5] Certain aspects of this disclosure relate to methods for determining corrected color values from image data obtained by a nucleic acid sequencer during a base recognition cycle, the sequencer comprising an image acquisition system, one or more processors, and a memory device. Such methods may be characterized by the following steps: (a) obtaining an image of a substrate (eg, part of a flow cell) containing a plurality of nucleobase readout sites; (b) measuring color values of a plurality of regions from the substrate image, (c) storing the color values in a processor buffer of one or more sequencer processors; (d) obtaining (returning) partial phasing corrected color values for a plurality of regions, wherein the partial phasing corrected color values were stored in the sequencer memory during the immediately preceding base recognition cycle; (f) determining the advanced phasing correction; and (f) determining corrected color values. In various embodiments, all of these operations are performed during a single base recognition cycle. In some embodiments, the methods also include applying adjusted color values to discriminate bases for multiple regions.

[6] During the sequencing process, regions are colored to represent the types of nucleic acid bases. The measured and stored color values may be intensity values or other magnitude values at a specific wavelength or range of wavelengths. In some embodiments, color values are determined from only two channels of the sequencer. In some embodiments, color values are obtained using four channels of a sequencer. Although the present disclosure focuses on correcting the phasing of color signals, these concepts are also applicable to other types of signals generated during the sequencing of clusters of nucleic acids having identical sequences. Examples of such other signals include radiation outside the visible spectrum, ion concentration, etc.

[7] In some embodiments, step (f) uses (i) color values in a processor buffer, (ii) partially phase-corrected values stored from the immediately preceding loop, and (iii) forward correction to determine corrected color values in step (f). phasing. In some embodiments, step (e) uses (i) partially phase-corrected color values stored during the immediately preceding base detection cycle and (ii) color values stored in a processor buffer to determine the advanced phasing correction in step (e).

[8] In some embodiments, the phasing advance correction includes a weighting factor (weight). In such embodiments, the operation of determining the corrected color values may include multiplying the weights by the color values of the plurality of regions measured from the substrate image.

[9] In some embodiments, the methods also include determining a phasing correction for the immediately following base recognition cycle. As an example, determining the phasing correction for the immediately following base discrimination cycle involves analyzing (i) the partially phasing corrected color values stored in the sequencer memory and (ii) the color values stored in the processor buffer. In some embodiments involving determining a phasing correction for the immediately following base recognition cycle, the methods also include (i) obtaining partially phasing corrected color values for the immediately following base recognition cycle by applying the phasing correction to the color values of a plurality of regions stored in a storage device sequencer; and (ii) storing the partially phase-corrected color values for the immediately following base discrimination cycle in the sequencer memory. In some embodiments, obtaining the partial phase-corrected color values for the immediately following base recognition cycle also includes summing (i) the multiple-region phase-corrected color values and (ii) the multiple-region color values from the substrate image measured in (b). In some embodiments, when storing partially phase-corrected color values for the immediately following base recognition cycle, the partially corrected color values are stored in sequencer memory segment buffers.

[10] In some embodiments, the methods are performed in real time while sequence reads are obtained by a nucleic acid sequencer. In some embodiments, the nucleic acid sequencer performs sequencing by synthesizing nucleic acids at multiple sites. In some embodiments in which the substrate contains a flow cell, the flow cell is logically divided into segments and each segment represents a region of the flow cell containing a subset of regions, which subset is captured in a single image by the imaging system.

[11] In some embodiments using such systems, in step (d) (retrieving partial phasing corrected color values for a plurality of regions), the partial phasing corrected color values were previously stored in segment buffers of the sequencer memory, wherein the segment buffers are configured to storing data representing images of individual segments on the substrate. In some embodiments, the storage capacity is about 512 Gigabytes or less, or about 256 Gigabytes or less. In some embodiments, for example, the storage device has a capacity that is half the capacity required to store the data contained in all segments in two flow cells. In some embodiments, the processing described herein saves at least about 50 Gigabytes; and in some embodiments, at least about 100 Gigabytes.

[12] In some embodiments, before performing step (a) (taking an image of the substrate), the methods also include placing (providing) reagents in the flow cell and allowing the reagents to interact with the sites, resulting in the sites having colors corresponding to the types of nucleic bases , during the base recognition cycle. In such embodiments, the method may also include, after step (f) (determining the corrected color values): (i) placing fresh reagents in the flow cell and allowing the fresh reagents to react with the patches, causing the patches to have colors corresponding to the types of nucleobases , for the next base recognition cycle; and (ii) repeating steps (a)-(e) of the next base recognition cycle. Such methods may also include creating a first processor thread to perform operations (a)-(f) for a base recognition cycle and creating a second processor thread to perform operations (a)-(f) for the next base recognition cycle. In some embodiments, the methods also include allocating a processor buffer and a second processor buffer, wherein the second processor buffer is used to determine the adjusted color values in (f).

[13] Certain other aspects of the disclosure relate to nucleic acid sequencers, which may be characterized by the following elements: imaging systems; storage device; and one or more processors configured or configured to: (a) obtain data representing an image of a substrate comprising a plurality of regions at which nucleic acid bases are read (at which regions appear, for example, colors corresponding to types of nucleic acid bases); (b) obtaining color values for a plurality of regions from the substrate image; (c) storing color values in a processor buffer; (d) retrieving partial phasing corrected color values for a plurality of sites for the base recognition cycle (the partial phasing corrected color values were stored in the sequencer memory during the immediately preceding base recognition cycle); (e) determining the advanced phasing correction; and (f) determining corrected color values from, for example, (i) color values in a processor buffer, (ii) partially phasing corrected values stored during the immediately preceding cycle, and (iii) phasing advance correction.

[14] Instructions or other configuration (embodiment) for determining the phasing advance correction may include a configuration for determining the phasing advance correction from (i) the partial phasing correction color values stored during the implementation of the immediately preceding base recognition cycle, and (ii) the values colors stored in the processor buffer.

[15] In some embodiments, the storage device is divided into a plurality of segment buffers, each of which is configured to store data representing one segment image on a substrate. In some embodiments, the storage device has a capacity of less than about 550 Gigabytes (in some examples, less than half the capacity required to store the data contained in all segments in two flow cells).

[16] Processors can be configured to perform the described operations in a variety of ways, such as by receiving executable machine-readable instructions. In some cases, processors are programmed using hardware-based software or specialized processing cores, such as digital signal processing cores. In various embodiments, the processor(s) are configured or configured to perform (and/or control) any one or more of the method operations described above.

[17] In some embodiments, the phasing correction functions disclosed herein can significantly reduce the cost of a sequencing instrument through more efficient use of a storage device (eg, random access memory (RAM)). In some embodiments, these phasing correction functions are used in the context of real-time analysis (RTA) on sequencing platforms.

[18] These and other features of the disclosure will be presented in more detail below with reference to the corresponding drawings.

BRIEF DESCRIPTION OF THE DRAWINGS

[19] In FIG. Figure 1 shows a block diagram of a sequencer with hardware for real-time analysis of image data obtained from nucleic acid clusters.

[20] In FIG. Figure 2 provides an illustration of dual-channel sequencing data to illustrate the concepts of phasing and phasing-advance.

[21] In FIG. 3 shows the architecture of a flow cell containing a plurality of segments, each of which contains a plurality of clusters.

[22] In FIG. 4 shows a data array containing value data for clusters in a segment or other displayed fragment of a flow cell; wherein these values may be light intensity values for each of two or more color channels.

[23] In FIG. 5 schematically illustrates a first processing configuration and a method for performing real-time phasing correction.

[24] In FIG. 6 is a flow diagram of a base recognition process in which the processor and memory configuration shown in FIG. 5.

[25] In FIG. 7 schematically illustrates a second processing configuration and a method for performing real-time phasing correction. This configuration reduces system storage requirements.

[26] In FIG. 8 schematically illustrates a third processing configuration and a method for performing real-time phasing correction. This configuration further reduces system storage requirements.

[27] In FIG. 9 is a high-level block diagram of the first few processing cycles that may be used with the processor and memory configuration shown in FIG. 8, and, in some embodiments shown in FIG. 7.

[28] In FIG. 10 shows a flow chart of processing cycles in which base recognition is carried out with fully corrected phasing. Such a cycle can be performed in the third and subsequent processing cycles when sequencing segment clusters.

[29] In FIG. 11 presents comparative data for phasing correction methods using an algorithm for using a reduced volume of the main storage device.

DETAILED DESCRIPTION

DEFINITIONS

[30] Numeric ranges include numbers that define a range. Each maximum numeric limitation set forth herein is intended to include each lesser numeric limitation as if such lesser numeric limitations were expressly stated herein. It is intended that each minimum numerical limitation set forth herein will include each higher numerical limitation as if such higher numerical limitations were expressly stated herein. Each numerical range set forth herein will include each narrower numerical range referred to within such broader numerical range as if all such narrower numerical ranges were expressly set forth herein.

[31] The headings set forth herein are not intended to limit disclosure.

[32] Unless otherwise defined herein, all technical and scientific terms used herein have the same meaning as commonly understood by those skilled in the art. Various scientific vocabularies that include the terms included herein are well known and available to those skilled in the art. Although any methods and materials similar or equivalent to the methods and materials described herein find use in the practice or testing of the embodiments disclosed herein, certain methods and materials are described.

[33] The terms defined below are described in more detail by reference to the full specification. It should be understood that this disclosure is not limited to the specific method, protocols and reagents described, as these may vary depending on the context of their use by those skilled in the art.

[34] The singular grammatical forms used herein are intended to refer to the plural unless the context clearly indicates otherwise. The term "set" refers to more than one element. For example, the term is used herein in reference to the number of reads to obtain a phased islet using the methods disclosed herein.

[35] The term “fragment” is used herein to refer to an amount of genome, chromosome, or haplotype sequence information in a biological sample that totals less than the amount of sequence information of one complete genome, one complete chromosome, or one complete haplotype as apparent from the context .

[36] The term “sample” as used herein refers to a sample, typically obtained from a biological fluid, cell, tissue, organ or organism, and containing a nucleic acid or mixture of nucleic acids containing at least one nucleic acid sequence to be sequenced. Such specimens include, but are not limited to, sputum/oral fluid, amniotic fluid, cerebrospinal fluid, blood, blood fraction (e.g., serum or plasma), fine needle biopsy specimens (e.g., surgical biopsy, fine needle biopsy, etc.). etc.), urine, saliva, semen, sweat, tears, abdominal fluid, pleural fluid, explant tissue in lavage fluid, organ culture and any other tissue or cellular preparation, or a fraction or derivative thereof, or isolation from them.

[37] Although the sample is often taken from a human (eg, a patient), samples can be taken from any organism that has chromosomes, including, but not limited to, dogs, cats, horses, goats, sheep, cattle, pigs, etc. A sample can be used directly when obtained from a biological source or after pre-treatment to modify the properties of the sample. For example, such pretreatment may include preparing plasma from blood, diluting viscous liquids, etc. Pretreatment methods may also include, but are not limited to, filtration, sedimentation, dilution, distillation, mixing, centrifugation, freezing, lyophilization, concentration, amplification, fragmentation of the nucleic acid, inactivation of unwanted components, addition of reagents, lysing, etc. If such pretreatment methods are applied to a sample, in such pretreatment methods the nucleic acid(s) of interest typically remain in the sample of interest, sometimes at a concentration proportional to the concentration in the untreated test sample (eg, specifically in a sample to which no such pre-treatment method(s) have been applied). Such “processed” or “prepared” samples are still considered biological “research” samples in the context of the methods described herein.

[38] The terms "polynucleotide", "nucleic acid" and "nucleic acid molecules" are used interchangeably and refer to a covalently linked sequence of nucleotides (i.e., ribonucleotides in the case of RNA and deoxyribonucleotides in the case of DNA) in which the 3'- The pentose position of one nucleotide is connected by a phosphodiester group to the 5' pentose position of the next nucleotide. Nucleotides include sequences of any form of nucleic acid, including, but not limited to, RNA and DNA molecules. The term "polynucleotide" includes, without limitation, single-stranded and double-stranded polynucleotide.

[39] Single-stranded polynucleotide molecules can be formed in single-stranded form, as DNA or RNA, or formed in double-stranded DNA (dsDNA) form (eg, genomic DNA segments, polymerase chain reaction (PCR) and amplification products, etc.). Thus, a single-stranded polynucleotide may be the coding or non-coding strand of a double-stranded polynucleotide. Methods for preparing single-stranded polynucleotide molecules suitable for use in the described methods using standard methods are well known in the art. The exact sequence of the primary polynucleotide molecules is generally immaterial to the disclosed embodiments and may or may not be known. The single-stranded polynucleotide molecules may be genomic DNA molecules (eg, human genomic DNA), including both intronic and exonic sequences (coding sequences), as well as non-coding regulatory sequences such as promoter and enhancer sequences.

[40] The nucleic acid described herein can be of any length suitable for use in the provided methods. For example, the target nucleic acids may have a length of at least 10, at least 20, at least 30, at least 40, at least 50, at least 75, at least 100, at least 150, at least at least 200, at least 250, at least 500, or at least 1000 kilobases (kilobases = thousand base pairs) or more.

[41] In the context of a flow cell or other sequencing substrate, the term "site" refers to the small area where sequencing is performed. In many embodiments, a region contains multiple, typically large, multiple copies of a single nucleic acid sequence from which sequencing data is derived. The sequence data obtained from a site may be “reads”.

[42] The term "polymorphism" or "genetic polymorphism" is used herein to refer to the presence of two or more alleles at a single genetic locus in the same population. Various forms of polymorphism include single nucleotide polymorphisms, tandem repeats, microdeletions, insertions, insertion-deletion mutations, and other polymorphisms.

[43] The term "base recognition" refers to the assigned base (nucleotide type) to sequence data for a specific position in a polynucleotide sequence. A recognized base can be output by the sequencer for each position in the nucleic acid being sequenced. Sometimes base recognition is characterized by the quality of recognition.

[44] The term "read" refers to the sequence read from a fragment of a nucleic acid sample. Typically, although not necessarily, a read represents a short sequence of adjacent base pairs in a sample. A read can be symbolically represented by a sequence of base pairs (like ATCG)) of a sample fragment. It may be stored in a storage device and processed accordingly to determine whether it matches a reference sequence or other criteria. The read can be obtained directly from the sequencing device or indirectly from stored sequence information related to the sample. In some cases, a read is a DNA sequence of sufficient length (eg, at least about 25 base pairs) that can be used to identify a larger sequence or region, for example, that can align to and be specifically associated with a chromosome, or a genomic region or genome.

[45] The term “next generation sequencing (NGS)” as used herein refers to sequencing methods that provide massively parallel sequencing of co-amplified molecules and single nucleic acid molecules. Non-limiting examples of next generation sequencing include sequencing by synthesis using reversible dye-labeled terminators and sequencing by ligation.

[46] The term “parameter” as used herein refers to a numeric value that characterizes a physical property or a representation of that property. In some situations, a parameter numerically characterizes a set of quantitative data and/or a numerical relationship between sets of quantitative data. For example, the mean and variance of the standard distribution approximation for a histogram are parameters.

[47] The terms “threshold” as used herein refer to any number used as a cut-off value to characterize a sample, nucleic acid, or fragment thereof (eg, a read). A threshold value can be compared to a measured or calculated value to determine whether the source of that value should be classified in a particular way. Threshold values can be determined empirically or analytically. The choice of threshold value depends on the level of confidence that the user will use in the classification. Sometimes they are chosen for a specific purpose (for example, to balance sensitivity and selectivity).

[48] Real-time analysis refers to the process and system in which data processing and analysis is performed in the background relative to data collection during DNA sequencing. An example of a real-time analysis system is described in US Pat. No. 8,965,076, which is incorporated herein by reference in its entirety.

CONTEXT FOR PHASING

Sequencing device

[49] In FIG. 1 is a block diagram of some of the components of a typical nucleic acid sequencer 100 or a system including such a sequencer. Specifically, system 100 includes a flow cell 101 and an image acquisition system 103, one or more processors 105 with one or more buffers 107, and a system memory (sometimes referred to as main storage) 109 containing a plurality of segment buffers 111. Typically, system storage 109 is located in a device that is not part of an integrated circuit containing any of one or more processors 105. In some embodiments, system storage is volatile storage, such as random access memory or RAM. for example, dynamic random access memory (DRAM), solid state drive, or hard disk drive.

[50] The flow cell and imaging system contain components made or configured in accordance with principles understood in the field of nucleic acid sequencing, and they will not be described in detail herein. Suitable image analysis systems and associated flow cells are used in nucleic acid sequencers, such as the MiSeq and HiSeq series sequencers manufactured by Illumina, Inc., San Diego, California. For further information, see US Patent No. 8,241,573, US Patent No. 9,193,996, and US Patent No. 8,951,781, each of which is incorporated herein by reference in its entirety.

[51] Typically, nucleic acid sequences suitable for use with the disclosed methods provide rapid and efficient simultaneous detection of multiple target nucleic acids. These may include fluidic components capable of delivering amplification reagents and/or sequencing reagents to one or more immobilized DNA fragments, the system comprising components such as pumps, valves, reservoirs, fluid lines, etc. A flow cell may be installed and/or used in an integrated system for detecting target nucleic acids. Exemplary flow cells are described, for example, in 2010/0111768 A1 (US) and No. 13/273,666 (US), each of which is incorporated herein by reference in its entirety. As shown for flow cells, one or more fluid components of the integrated system can be used for both the amplification method and the detection method. For example, one or more liquid components of an integrated system can be used for an amplification method and for delivering sequencing reagents in a sequencing method. Alternatively, the integrated system may comprise separate fluid systems for carrying out the amplification methods and for carrying out the detection methods.

[52] For purposes of the present disclosure, it is sufficient to understand that the flow cell first receives and immobilizes or otherwise fixes the nucleic acid sample to be sequenced and then exposes it to various reagents associated with the sequencing process. In some embodiments, the sequencing process is a sequence of the synthesis process, although other sequencing technologies may be used.

[53] The imaging system 103 includes optical components, such as fluorescence excitation components (e.g., a laser and associated mirrors and lenses) for illuminating portions of the flow cell in which sequencing occurs, and image capture components for capturing fluorescence images of portions of the flow cell having many areas. The data produced by the imaging system contains information that can be used to determine which nucleotide is being read at any given site in any given sequencing run.

[54] To perform real-time analysis, the sequencer 100 typically includes embedded processors and storage that interpret and store image data from the imaging system 103. Examples of suitable processors for the sequencer include processors in the Intel Xeon E5 class. Typically, processor 105 includes a plurality of buffers 107 that temporarily store image data acquired during one image acquisition cycle. In the illustrated embodiment, processor buffers are allocated in system storage. This processor buffer may be associated with a specific processor thread created to analyze the flow cell region image data during real-time analysis. In some embodiments, the image data analyzed as a stream is data from a single segment (described below) captured during one image acquisition cycle. In some embodiments, the buffer can store approximately 400 Gigabytes of data. In the context of this document, a thread is an ordered sequence of instructions that tells the processor what operations to perform. The instructions configure the processor using executable machine code selected from a specific set of machine language instructions, or "native instructions" built into the hardware processor.

[55] The machine language instruction set or native instruction set is known to the hardware processor(s) or central processing units and is essentially built into them. It is the “language” by which the system and application program communicate with hardware processors. Each native instruction is a discrete code recognized by the processing architecture and which can specify specific registers for arithmetic, address, or control functions; specific storage areas or offset values; and the specific addressing modes used to interpret the operands. More complex operations are created by combining these simple commands of their own, executed sequentially or otherwise as specified in control flow commands.

[56] System memory 109 includes a plurality of segment buffers 111, each of which is configured to store a portion of image data acquired from the flow cell during one image acquisition cycle. The segment buffers in this example are so named because they are configured to store image data of a single segment. As explained in more detail below, a segment is an area of a flow cell that can be captured in a single image acquired during a single imaging cycle. Segment buffers 111 are configured to store image data for a longer period of time compared to processor buffers 107. In some embodiments, segment buffers 111 store image data for at least two image acquisition cycles. Although this application describes buffers that buffer flow cell segment data, the disclosed embodiments are not limited to buffers that store this amount of data. Unless otherwise indicated or unless the context clearly indicates otherwise, references to “segment buffers” are intended to include any type of buffer storing image data of a portion of a flow cell, said image data being processed as a whole as described herein.

[57] To perform radix recognition, one or more processors 105 utilize data received from system storage 109 and data stored in processor buffers 107 . Typically, one base recognition is performed on one site during one image acquisition cycle.

[58] As shown in the figure, one or more processors 105 and main storage device 109 share data in two directions. In addition, one or more processors 105 receive image data from the image acquisition system 103. In some embodiments, the imaging system 103 obtains data from the flow cell 101 by driving sequencing sites in the flow cell 101 and receiving optical signals from those sites. In some embodiments, the signal received by the imaging system 103 is a fluorescence signal produced when the system 103 illuminates the flow cell 101 with light of appropriate wavelengths. In such embodiments, a fluorescence signal is obtained in the form of intensity values for a plurality of colors.

[59] The concept of a cycle is used throughout this disclosure. One sequencing run involves reading one nucleotide at each one or more regions captured in the image. Reading is called base recognition. In various embodiments described herein, both base recognition and image fixation are performed in a single computational cycle—in the context of the operation of the processor(s) and storage device—but for different nucleotides, with the base recognition being delayed relative to image fixation at sequences of read or recognized nucleotides. For example, in one computational cycle, one or more processors perform base recognition for a nucleotide in sequencing cycle n and simultaneously perform an image capture for the nucleotide in sequencing cycle n+1. Thus, in one computational cycle, the sequencer (a) stores and processes unmodified image data for nucleotides in sequencing cycle n+1 and (b) performs base discrimination for nucleotides in sequencing cycle n. The use of processor buffers and segment buffers in this round-by-cycle processing will be described in more detail below.

General information about phasing

[60] Multiple copies of a nucleic acid molecule having the same sequence (possibly with limited variations unintentionally introduced during sample processing) are simultaneously analyzed at a specific location in a flow cell or other support. Sufficient copies are used to generate sufficient signal to ensure reliable base recognition. A set of nucleic acid molecules in a site is called a cluster. In some cases, the unsequenced cluster contains only single-stranded nucleic acid molecules.

[61] Phasing is an off-target artifact that results from sequencing multiple nucleic acid molecules in a cluster. Phasing is the degree to which signals, such as fluorescence from individual molecules in a cluster, lose synchronization with each other. Often the term “phasing” is used to refer to interference from some molecules that are lagging, and the term “pre-phasing” (advanced) is used to refer to interference from other molecules that are leading. Together, phasing and phasing forward determine how well the sequencing device performs and the chemical analysis is performed.

[62] Lower values are better. Values of 0.10/0.10 mean that 0.10% of the molecules in the cluster are lagging and 0.10% are leading in each base recognition cycle. In other words, 0.20% of the true signal is lost in each cycle and will therefore introduce noise. In another example, 0.20/0.20 means that 0.4% of the true signal is lost per cycle, in which case after 250 cycles (without correction) the noise will be equal to the signal.

[63] The real-time analysis component of the sequencer can detect phasing and phasing advance to apply the correct level of phasing correction as the sequence progresses. This is done by artificially forcing a signal into or out of each sequencer channel based on base detections before or after the current cycle.

[64] Previously, phasing and phasing advance were assessed over a specified number of cycles (eg, the first 12 cycles of each read) and then applied to all subsequent cycles. Some modern sequencers use an algorithm called empirical phasing correction to optimize phasing correction on each cycle by adjusting the correction range to select the range that provides the highest fidelity (signal purity). Although empirical phasing correction provides improved performance, it requires more computational resources.

[65] In conventional sequencers, each base has a unique fluorescent dye color; for example, green for thymine, red for cytosine, blue for guanine, and yellow for adenine. To obtain base discrimination information, a four-channel sequencer acquires four images of a segment or other portion of the flow cell. Some sequencers now have only two channels and thus obtain only two images of the same section of the flow cell. A two-channel sequencer uses a mixture of dyes for each base and uses red and green filters for the two images. In the two-channel sequencer example, clusters visible in the red or green images are interpreted as C and T bases, respectively. Clusters visible in both the red and green images are labeled A bases, and unlabeled clusters are defined as G bases.

[66] In FIG. Figure 2 shows phasing during sequencing of a nucleic acid cluster having the sequence . . . ACGTAAG. . . . As shown in the figure, during a base recognition cycle for the first base G, 98.4% of the fluorescence signal is generated by sequences currently generating a signal for G, while 1.5% of the fluorescence signal is generated by sequences currently generating a signal for the preceding base C, and 1.1% of the fluorescence signal is generated by sequences currently generating the signal for the following base T. The signal for the preceding base C is contributed by phasing, and the signal for the following base T is contributed by advanced phasing.

[67] The phasing correction to recognize this nucleotide base G is depicted in the graph on the right side of FIG. 2. As shown for a two-channel sequencer, the fluorescence signal can be represented on a two-dimensional plot with the maximum intensity signal on the “green axis” representing the T base, the maximum intensity on the “red axis” representing the C base, the maximum intensity midway between the axes representing base A, and a minimum intensity on both axes representing base G. In the absence of phasing errors, the signal for base G should have zero intensity on both the red and green axes. Instead, with the phasing error discussed, the fluorescence signal is characterized by a slight increase in intensity on both the green and red axes. In this example, the phasing advance correction causes the signal intensity to decrease to zero on the green axis, and the phasing correction causes the signal intensity to decrease to zero on the red axis. Similar correction can be made in base recognition for T, C, and A bases.

Segments and flow cells

[68] As explained above, a flow cell contains multiple sites at which sequence information is collected. In some embodiments, each region of the flow cell contains a cluster of single-stranded nucleic acids having the same sequence. A single image used in real-time sequencing may contain millions of such clusters. A typical flow cell is so large that hundreds or even thousands of individual images are required to cover its entire area. In some embodiments, the processor and associated memory used for real-time analysis processes all of these images in an ongoing manner to perform base discriminations for a single cycle. In some embodiments, the processor and memory simultaneously process all images acquired in two or more flow cells during one base recognition cycle. In fig. Figure 3 schematically shows the flow cell architecture used in some sequencers manufactured by Illumina, Inc. In the example shown, the sequencer performs simultaneous base sensing on two flow cells, Flow Cell 1 and Flow Cell 2. In some embodiments, each flow cell has sequencing sites on each of two surfaces, a top surface and a bottom surface. In such cases, the sequencer displays both the top and bottom surfaces during each base recognition cycle. As shown in FIG. 3, each surface of the flow cell contains four tracks, L1, L2, L3 and L4; Of course, another number of them is possible. Each track of each surface can have multiple sections called stripes. Each strip is in turn divided into many segments. For example, one strip may contain approximately 120 segments. For two flow cells, each flow cell having two surfaces, each with four lanes, each lane containing six lanes, and each lane containing 120 segments, data from several thousand segments per cycle must be analyzed. In various embodiments, each segment image (or other image from a portion of the flow cell) utilizes one processor thread. In some embodiments, a sequencer that uses a flow cell having the architecture depicted in FIG. 3, processes 8000 or more segments in each base recognition cycle. In such cases, the real-time processing logic will use 8000 or more processor threads in each radix cycle.

[69] Data from one segment captured during one cycle may be stored in a storage device as an array, with each entry in the array representing a color value for each channel of a particular cluster in the segment. The array for a two-channel configuration is shown in FIG. 4. As an example, a color intensity detector can generate 400 to 1500 signals for each channel. The segment buffer in the system storage device is configured to store all the information in the array, in other words, the color values of all clusters on the segment in one base recognition cycle. The processor buffer may similarly be configured to store all information in an array.

Phasing process

[70] Significant memory load when analyzing real-time sequence data is due to the need for phasing correction, so two or three intensity cycles for a cluster must be stored for each segment for the full cycle length. On the Illumina HiSeqX with a 700 nm flow cell, this takes up 73 Gigabytes of storage. This load is large enough that most of the data (on this platform) fits into the cache on the SSD.

[71] As explained above, phasing correction provides adjustments to image intensity values to eliminate phase alternation of certain nucleic acid bases in a cluster. To do this, phasing correction begins with the measured cluster color intensities (or other signals measured in the sequencing method) for the current base discrimination cycle and adds or subtracts a correction value using the measured intensities from the previous base recognition cycle and/or using the measured intensities from subsequent base recognition cycle. In various embodiments, the phasing corrected intensity value for performing base discrimination is determined using the expression shown at the bottom of FIG. 5. As shown in the figure, the phasing-corrected intensity values for the current base recognition cycle in the image are equal to the measured intensity values for the current base recognition cycle minus the product of the first coefficient and the measured intensity values in the immediately preceding base recognition cycle and minus the product of the second coefficient and measured intensity values in the immediately following base recognition cycle:

Corrected intensity = -a.In-1 + In - b.In+1

where In-1, In and In+1 represent the intensity values of the clusters in the segment in the immediately preceding base recognition cycle, in the current base recognition cycle and in the next base recognition cycle. The coefficients a and b are, respectively, the phasing and phasing advance coefficients (sometimes called weights). They may be recalculated for each base recognition cycle for the segment.

[72] Returning to FIG. 2, the measured intensity value for the third base in the imaged sequence (for one cluster in the image) is shown as a point on the graph on the right side of FIG. 2. The advanced phasing correction of this measured intensity value is represented by a vertical arrow from the measured intensity value to the horizontal axis. In the expression for phasing-corrected intensity values, said phasing advance correction is represented by the product of the coefficient b and the intensity value measured for the next sequential base discrimination cycle. In addition, the measured intensity value is corrected using phasing correction, represented by a horizontal arrow in the graph. This phasing correction is accomplished by subtracting from the measured intensity value the product of the coefficient a and the measured intensity value for the immediately preceding base recognition cycle. The coefficients a and b can be determined in a variety of ways, but in many embodiments they are recalculated for each base detection cycle. A description of methods for determining the coefficients to be used in phasing correction is contained in International Patent Application Publication Number WO2015/084985, Belitz et al., published June 11, 2015, which is incorporated herein by reference in its entirety.

[73] In some embodiments, the phasing algorithm determines the phasing coefficients empirically by maximizing the aggregate confidence (or similar metric) of the cluster intensity data during the base recognition cycle. One implementation of the algorithm iterates over all or many of the phasing coefficients and determines which ones provide the best results. For example, a phasing algorithm may optimize a and b in each cycle using a pattern search using an optimality criterion that counts the number of clusters that fail the validity filter. Thus, a and b are chosen to maximize data quality.

[74] In some embodiments, the determination of phasing coefficients occurs as an ongoing analysis throughout the sequencing run (eg, during read generation). As a result of this approach, inaccurate phasing estimates obtained on earlier cycles will not have a negative impact on later cycles.

[75] In some methods, the confidence of a cluster intensity value is determined as a function of the relative distances to the Gaussian centroids for other cluster intensity values determined for the same base discrimination cycle. The centroids perfectly match the expected locations of the A, T, C and G intensities for the two channels (see Fig. 2) assuming a two-channel system is used. In some embodiments, the confidence may be calculated using the following expression:

reliability = 1 - D1 / (D1 + D2),

where D1 represents the distance to the nearest Gaussian centroid and D2 represents the distance to the next closest centroid. Using this approach, when the average confidence (quality) of the intensity values is maximized, the correct values of a and b are selected. Once these values have been determined, correction can be applied to all values in the cluster and base recognition can be directly performed. Methods for fitting Gaussian distributions for a two-channel data set are described in international patent application publication number WO2015/084985, previously incorporated by reference.

[76] In some embodiments, phasing correction is calculated at nearly every cycle during sequencing. In some embodiments, phasing correction is calculated at each cycle during sequencing. In some embodiments, a separate phasing correction is calculated for different locations of the imaged surface in the same cycle. For example, in some embodiments, a separate phasing correction is calculated for each individual track of the imaged surface, eg, an individual flow cell track. In some embodiments, a separate phasing correction is calculated for each subset of tracks, for example, a display strip in a flow cell track. In some embodiments, a separate phasing correction is calculated for each individual image, such as, for example, each segment. In some embodiments, a separate phasing correction is calculated for each segment in each cycle.

[77] As readouts become longer, higher order terms may become more important in phasing correction. Thus, in certain embodiments, an empirical second-order phasing correction may be calculated to correct for this. For example, in some embodiments, the method includes second order phasing correction, defined as follows:

I(cycle) = -a * I(cycle - 2) - A * I(cycle - I) + I(cycle) - B * I(cycle + 1) - b * I(cycle + 2)

where I represents the intensity, a, A, B and b represent the first and second order terms for phasing correction. In particular embodiments, the calculation is optimized over a, A, B, and b.

[78] In FIG. 5 schematically shows the processing configuration and method for performing real-time phasing correction. In the illustrated embodiment, processor 502 creates a new processing thread 503 when the processor is called to perform base recognition in clusters in an image, such as a segment image. A new thread can be generated for each base recognition cycle for each segment. In the illustrated embodiment, processor 502 ensures that one processor buffer 505 is available during each radix cycle for a segment (and assigned processing thread). The processor buffer temporarily stores intensity values that the processor uses during computation to perform phasing correction for the current base discrimination cycle n. In the illustrated embodiment, the processor interfaces with a system memory 507 containing three buffers, one each for storing image data captured for a particular base recognition cycle. When using the flow cell architecture shown in FIG. 3, each buffer stores image data for clusters of one segment; thus, these buffers are called segment buffers. Of course, for other flow cell architectures and/or imaging systems, these buffers may store more or less cluster data. For convenience, the term “segment buffer” will be used in this description. Each segment buffer stores data for one segment (or other portion of the flow cell) captured during one base sense cycle. The image data may be represented as a data array, such as that shown in FIG. 4.

[79] As shown, system memory 507 includes a segment buffer 509 that temporarily stores intensity values for the immediately preceding base sense cycle (compared to the current base sense cycle being processed by the processor), a segment buffer 511 that stores intensity values measured for the current base sensing cycle, and a segment buffer 513 that stores intensity values for the immediately next base sensing cycle. Again, each of the segment buffers 509, 511 and 513 contains data measured for one segment for one base discrimination cycle n.

[80] As shown, thread 503 uses intensity values stored in each of segment buffers 509, 511, and 513 during one base detection cycle. The intensity values are sequentially loaded into processor buffer 505 and processed to implement a calculation according to the phasing correction expression presented at the bottom of FIG. 5. After the base recognition process is completed, as shown in the processor and memory configuration in FIG. 5, the processor buffer contains the adjusted intensity values used to perform phasing-corrected base discrimination.

[81] In FIG. 6 is a flow diagram of a base recognition process in which the processor and memory configuration shown in FIG. 5. As shown in FIG. 6, process 601 initiates a new base detection cycle by creating a processor thread and allocating a processor buffer for that thread. See process block 603. The processor then extracts intensity data from an image of a flow cell segment (or other suitable portion of the flow cell) acquired concurrently with the current processing cycle. In the illustrated implementation, the captured image and corresponding intensity values are the primary intensity values for the next sequential base detection cycle, not the current base detection cycle (the current processing iteration). In other words, in the current processing cycle, base recognition is performed on the image data collected in the immediately preceding processing cycle. Thus, as shown in block 605 of process 601, the extracted intensity values are indicated by the notation In +1, where n represents the current base recognition cycle. In other words, the processing loop (i) performs base recognition for base recognition cycle n and (ii) commits image data for base recognition cycle n+1.

[82] The newly extracted intensity data, which may be provided in the form of an array as shown in FIG. 4 are stored in an available segment buffer in system storage (eg, segment buffer 513). In some embodiments, said segment buffer is a buffer that stores intensity data that was previously used but is no longer needed for base recognition.

[83] In the current processing cycle, process 601 also retrieves intensity data stored during the computation cycle preceding the current computation cycle. See process block 607. The extracted intensity data is related to the current base recognition cycle and is labeled In. The extracted intensity data is obtained from a corresponding segment buffer, such as system storage segment buffer 511, as shown in FIG. 5.

[84] In addition, process 601 retrieves intensity data that was stored two cycles prior to the current base discrimination cycle. See process block 609. As an example, with reference to FIG. 5, such intensity data may be obtained from system storage segment buffer 509. The array of intensity values extracted in operation 609 is denoted In -1.

[85] Although operations 605, 607 and 609 are shown to occur sequentially, the specified order of operations is flexible and the process can be implemented such that any order is acceptable as long as it is consistent with base recognition, which includes phasing correction.

[86] After retrieving the intensity values for the current base detection cycle (process block 607) and the intensity values for the immediately preceding base detection cycle (processing block 609), the processor has all the intensity values required to perform phasing correction. The processor performs phasing correction by first determining the leading phasing correction weight b and the phasing correction weight a for the current base recognition cycle. See process block 611 which illustrates that this can be accomplished using the extracted intensity values for the next base sensing cycle along with the intensity values for the current and immediately preceding base sensing cycles. Then, using the phasing correction and phasing advance weights, the processor calculates the phasing corrected intensity values for the current base recognition cycle, as shown in process block 613. The adjusted values refer to the clusters in the segment in question. The calculation may use the expression shown in block 613. Using the phasing corrected intensity values, the processor performs discrimination during the current base recognition cycle, as shown in process block 615.

[87] At this point, processing for the current base recognition cycle is completed and the next iteration of base recognition can be performed. Deciding whether to perform another round of base discrimination is depicted at block 617, which determines whether there are any additional nucleotides to be sequenced in the clusters of the segment in question. If they are absent, the process terminates as shown at block 619. If present, control of the process is transferred to process block 621, where the processor increments the loop counter. This allows the intensity values for the current base sensing cycle In to be effectively indexed with respect to the intensity values for the immediately preceding base sensing cycle In-1. At the same time, the intensity values for the immediately next basic (In+1) base discrimination become the intensity values for the current base discrimination cycle (In). These increments are performed on the indexes applied to the intensity data stored in the segment buffers.

Phasing process (with reduced main storage requirements)

[88] The approach shown in FIG. 5 and 6 can operate normally provided that the sequencer and the associated real-time analysis system are not limited in terms of storage capacity. However, given the volume of data that must be processed in some modern sequencers, such as those used to perform whole genome sequencing, the storage capacity may not be sufficient, particularly at a commercially viable cost. Thus, storing three times the amount of data corresponding to a complete image of the flow cell (or flow cells) during a base sensing cycle can lead to severe resource constraints.

[89] An algorithm for phasing, for example, shown in FIG. 5 and 6 is a significant benefit for real-time analysis as it significantly improves sequencing results, particularly on non-standard samples, such as those with low genetic diversity. However, the storage load increases as the performance of next-generation sequencing systems increases. The following embodiments reduce memory load by using phasing weights derived from data that has already been partially corrected for phasing. The phasing and phasing-advance weights can be obtained independently and still provide high-quality sequencing results. In some examples, the main storage requirements include half the volume required to store the data contained in all segments in two flow cells.

[90] In some embodiments, the processor and memory for phasing-corrected base recognition are configured to reduce system storage requirements. One example of this is shown in FIG. 7. The intensity values are adjusted as described above, for example, the phasing and phasing advance weights are calculated and applied to the immediately preceding and immediately following cycles. However, in the example shown in FIG. 7, system memory 707 uses only two segment buffers for phasing correction: segment buffer 709 and segment buffer 711. In this example, processor 702 uses processing thread 703, for which, unlike the example shown in FIG. 5, two associated processor buffers are used: processor buffer 705 for storing and manipulating intensity values retrieved from memory 707, and processor buffer 706 for storing and using intensity values In+1 of newly captured images. In the example shown, processor buffers are allocated in main storage, but this is not always necessary. In some embodiments, processor buffers are allocated in another physical storage device or even on a processor chip.

[91] Replacing segment buffers with processor buffers can significantly reduce overall memory requirements. When using multiple processors and/or multithreading, multiple processors process data from multiple segments. For example, the number of segments in a system may be on the order of 1000-2000, while the number of processors processing all these segments is approximately twenty. Theoretically, such a system can reduce storage space usage by approximately 50 times. In some embodiments, the reduction is on the order of 20 times.

[92] In this embodiment, the intensity values obtained from the segment images in the current processing cycle (In+1) are stored locally on the processor and are used to calculate the phasing and phasing advance weights and subsequently perform base recognition. In some implementations, the most recently received intensity values (In+1) are stored in a segment buffer in system storage 707 only after this process has completed.

[93] In some embodiments, the processor and system storage are configured as shown in FIG. 8. As with the processor/memory configuration shown in FIG. 7, processor 802 uses processing threads 803, each of which is associated with two processor buffers: processor buffer 805 for temporarily storing intensity values from system storage 807 (segment buffer 811) and processor buffer 806 for temporarily storing intensity values received during the current cycle (In+1) processing. To ensure that this configuration is effectively and efficiently applied to the intensity values stored in segment buffer 811, it is necessary to partially perform phasing correction. Examples of ways to do this are described below. Processor buffer 705 in FIG. 7 and processor buffer 805 in FIG. 8 load intensity values from main storage and then use them to generate corrected intensity values used for base discrimination. In the example shown, processor buffers are allocated in main storage, but this is not always necessary. In some embodiments, processor buffers are allocated in another physical storage device or even on a processor chip.

[94] In FIG. 9 is a high-level view of a process 901 that may be used with the processor and storage configuration shown in FIG. 8, and, in some embodiments shown in FIG. 7. As shown in FIG. 9, the first and second processing cycles do not use all the information to perform full phasing correction in the clusters displayed in the segment. However, in the first cycles, phasing is not a significant problem.

[95] To perform full phasing correction, the sequencer requires three consecutive image acquisition cycles. In the first round of processing, the sequencer does not perform base recognition; it only stores the intensity data for the next processing, i.e. the cycle in which the first base recognition will be performed.

[96] As shown, process 901 begins at process block 903, which creates a thread for the first processing cycle. The commands in this flow control the extraction of intensity data from the cluster image during the first sequencing cycle (I1), that is, the cycle during which the first nucleotides of the clusters are read. See process block 905. The image data is stored in a segment buffer in the system storage device. At this point, the first processing cycle is essentially completed.

[97] The process continues at process block 907, where a new thread is created in preparation for the second round of processing. In this process, the first and second processor buffers are allocated for the second processing cycle. See block 907. Process blocks 907, 909, 911, 913, 915, 917, 919, 921 and 923 are executed during a second processing loop, which is performed using the thread and processor buffers generated at process block 907.

[98] As shown, the processor extracts intensity data from the image for the next base detection cycle (I2) and stores this data in the first processor buffer. See process block 909. Then, during the second processing cycle, the processor retrieves the intensity data stored in the segment buffer during the first processing cycle, which represents the intensity data for the current base recognition cycle (I1). See block 911. Using the intensity data collected during the first and second processing cycles, the processor may calculate the phasing advance weight b for the current base recognition cycle (ie, the first base recognitions in the reads). See process block 913. Using the intensity values for the first two rounds and the phasing advance weighting factor, the processor calculates adjusted intensity data values for the second round (I2) of base recognition. The adjusted intensity data values may be stored in a second processor buffer. See process block 915. The processor then performs base discriminations for a second base discrimination cycle using the adjusted intensity data values obtained at block 915. See process block 917.

[99] At this point, the sequencing process is ready to begin preparing for the next round of base recognition. It begins at process block 919 by determining the phasing correction weight a using the intensity data (I2) of the next (or second) base discrimination cycle and the current base recognition cycle data (I1) that has been stored in the segment buffer. Using the phasing correction weight a, the processor then calculates the phasing corrected (but not forward phasing corrected) intensity data values from the current uncorrected intensity data (I2) extracted during this second round of processing and the intensity data values for the first loop (I1 ) processing according to the expression given in process block 921. This results in an array of partially corrected intensity values (I2(partially corrected)) for the second round of base recognition. The sequencer will have to wait until the next processing cycle to perform advanced phasing correction. However, at this point most of the computation is complete and the array data for one image can be stored in a segment buffer for use in the next round of base recognition. To this end, the processor stores the phasing corrected (but not forward phasing corrected) intensity data in the segment buffer (such that I2 (partially corrected) replaces I1 in the segment buffer). See process block 923.

[100] At this point, the first and second processing cycles are completed and base recognitions are performed for the first base recognition cycle, which is the second processing cycle. Subsequent base recognition cycles can be performed with full phasing correction as described in FIG. 10. See process block 925.

[101] In FIG. 10 illustrates the sequence of operations that can be performed during a processing cycle in which phasing-corrected base recognition is performed. Such a cycle can be performed in the third and subsequent processing cycles when sequencing segment clusters. In some embodiments, the process flow depicted in FIG. 10 corresponds to process block 925 shown in FIG. 9.

[102] As shown in the figure, the process begins by allocating a thread and its associated first and second processor buffers. See process block 1003. Next, the processor extracts intensity data values from the image for the next round (In+1) of base recognition and stores these values in the first processor buffer. See process block 1005. Concurrently, the processor retrieves the partially corrected intensity data values that were stored during the previous base recognition cycle (as a non-limiting example, I2(partially corrected) in the embodiment shown in FIG. 9, or In - a(In-1)). These values now represent the intensity values for the current base recognition cycle (In). They were previously stored in the system storage segment buffer and are now retrieved from it. See process block 1007. With the partially corrected intensity data values for the current base recognition cycle for which the phasing has been adjusted, the processor only needs to perform a forward phasing correction to complete the intensity data correction and perform the necessary base recognitions for the current base recognition cycle. To this end, the processor determines the phasing advance correction weight b for the current base recognition cycle. To accomplish this, the processor uses the extracted intensity data it has just extracted from the image data for the next cycle (In+1) along with the previously partially corrected intensity data for the current base detection cycle. Recall that this partially corrected data has just been retrieved from the segment buffer. The partially corrected intensity data can be represented by the following expression: In - a(In-1). See process block 1009.

[103] Given the phasing advance correction weight b calculated for the current base recognition cycle, the processor has all the necessary data to compute a fully phasing corrected intensity data array for the current base recognition cycle (In). The calculation is carried out as shown in process block 1009. The resulting fully corrected intensity data values are stored in a second processor buffer. See process block 1011. The processor then performs base discriminations for the current base recognition cycle using the adjusted intensity data values stored in the second processor buffer. See process block 1013.

[104] The current processing cycle can begin preparation for the next base recognition cycle, which will be performed during the next processing cycle. In the illustrated embodiment, the processor determines the phasing correction weight a for the next base recognition cycle using the intensity data available for the current base recognition cycle. See process block 1015. Recall that the intensity data of the next base discrimination cycle was retrieved and stored in the first processor buffer at process operation 1005. Partially corrected intensity values for the current base discrimination cycle were retrieved from the segment buffer in order to perform ongoing base discriminations. The same partially corrected intensity values are now used to calculate the phasing correction weight a for the next base discrimination cycle. Now, using the calculated phasing correction weight for the next base recognition cycle, the processor calculates phasing corrected (but not advance phasing corrected) intensity data values, as shown in process block 1017. The processor then stores these phasing-corrected intensity data values for the next base detection cycle in a segment buffer. See process block 1019.

[105] Prior to this invention, it was believed that base recognition accuracy would deteriorate when phasing advance weights were determined from phasing corrected intensity values. However, the results reported herein indicate that there are few or no inaccurate results. In some embodiments, the image data is compressed (eg, through lossy compression) and even the data is compressed with partial phase correction. In both cases it was demonstrated that compression could be performed without loss of accuracy. For example, a non-compressed implementation uses two floating buffers for each segment (the floating buffer size is 4 bytes). When compression is used, the implementation uses a one-byte buffer, thus using 4 times less memory.

[106] At this point, the current processing cycle is essentially complete, so the processor determines whether any more cycles need to be performed when sequencing the clusters of the current segment. See decision block 1021. If no more bases need to be read from the clusters, the process ends and no further processing cycles are performed. However, if one or more additional sequencing cycles are required, process control is transferred to process block 1023, where the processor increments the current cycle number, at which point the partially corrected intensity data values stored in the segment buffer become current; that is, they become values for a new base recognition cycle. Process control then returns to process block 1003, where the next processing cycle begins.

EXAMPLE

[107] As explained above, some embodiments reduce memory load by using phasing weights derived from data that has already been partially phasing corrected. However, it was not realized that phasing and phasing advance weights could be obtained independently and still provide high-quality sequencing results. From the example shown in FIG. 11, it is clear that this is possible.

[108] As shown in the figure, two comparisons were made, each using a basic process (eg, the process shown in FIGS. 5 and 6) and a new process that was optimized to reduce main storage requirements (eg , the process shown in Figures 8 and 10). The same sequencer and sample were used in each comparison. Specifically, the Illumina HiSeqX instrument has been converted to use a 2-dye chemistry. The sequencer output images were saved and both phasing algorithms were tested on the same sequencing images, allowing for a fully controlled test. The Performance by Cluster column indicates the performance provided by the sequencer; the "% Aligned" column indicates the number of clusters that are successfully aligned to the reference genome, and the "% Error Rate" column indicates the average frequency of sequence errors caused by the software compared to the reference genome.

[109] As can be seen from the sequencing results, the memory efficient phasing algorithm is comparable to the baseline algorithm. In this example, the use of a memory efficient process resulted in an increase in error rate of approximately 3%, which was offset by a reduction in the usable size of the main storage device (estimated at 420 Gigabytes to 340 Gigabytes in some embodiments).

SEQUENCING METHODS

[110] As stated above, the present disclosure relates to the sequencing of nucleic acid samples. Any of a variety of sequencing technologies may be used using one or more channels of base recognition information, in particular optical channels. In particular, methods may be used in which nucleic acids are attached at fixed locations in an array (eg, in a cluster) and in which the array is repeatedly imaged. In particular, embodiments may be used in which images are obtained in different color channels, for example corresponding to different labels used to distinguish one type of nucleotide base from another. In some embodiments, the process of determining the nucleotide sequence of a target nucleic acid may be an automated process. Some embodiments include sequencing by synthesis (SBS) methods. Although this document emphasizes sequencing by synthesis, other sequencing technologies may be used.

[111] In many embodiments, synthetic sequencing methods involve enzymatic extension of a nascent nucleic acid chain by iterative addition of nucleotides and a non-coding strand. In traditional sequencing-by-synthesis methods, a single nucleotide monomer can be provided for each delivery of a target nucleotide in the presence of a polymerase. However, according to the methods described herein, more than one type of nucleotide monomer can be provided to a target nucleic acid in the presence of a polymerase upon delivery.

[112] Synthetic sequencing can use nucleotide monomers that have a terminator component or that do not have a terminator component. Methods using nucleotide monomers that lack terminators include, for example, pyrosequencing and sequencing using γ-phosphate labeled nucleotides. In methods using nucleotide monomers that lack terminators, the amount of nucleotides added in each cycle is typically variable and depends on the template sequence and the method of delivery of the nucleotides. For sequencing-by-synthesis methods that use nucleotide monomers with a terminator component, the terminator may be substantially irreversible under the sequencing conditions used, as in the case of traditional Sanger sequencing, which uses dideoxynucleotides, or the terminator may be reversible, as in the case of sequencing methods , developed by Solexa (now Illumina, Inc.).

[113] Synthetic sequencing methods may use nucleotide monomers that have a tagged component or do not have a tagged component. Accordingly, switching events can be detected based on a characteristic of the label, such as fluorescence of the label; characteristics of the nucleotide monomer, such as molecular weight or molecular charge; a by-product of nucleotide incorporation, such as pyrophosphate release; or the like. In embodiments in which two or more different nucleotides are present in the sequencing reagent, the different nucleotides may be distinguishable from each other or, Alternatively, two or more different tags may be indistinguishable by the detection methods used. For example, different nucleotides present in a sequencing reagent may have different labels and can be distinguished using suitable optical devices, as exemplified by sequencing methods developed by Solexa (now Illumina, Inc.).

[114] Some embodiments include pyrosequencing methods. Using pyrosequencing, the release of inorganic pyrophosphate (PPi) is detected when specific nucleotides are incorporated into the nascent chain (Ronaghi, M., Karamohamed S., Pettersson, B., Uhlen, M., Nyren, P. (1996) “Real-time DNA sequencing using detection of pyrophosphate release." Analytical Biochemistry 242(1), 84-9; Ronaghi, M. (2001) "Pyrosequencing sheds light on DNA sequencing." Genome Res. 11(1), 3-11; Ronaghi, M. ., Uhlen, M., Nyren, P. (1998) "A sequencing method based on real-time pyrophosphate." Science 281(5375), 363; US Patent No. 6,210,891; US Patent No. 6,258,568 and US Patent No. 6,274,320, disclosure which are incorporated herein by reference in their entirety). In pyrosequencing, the released PPi pyrophosphate can be detected by direct conversion to adenosine triphosphate (ATP) by ATP sulfurylase, and the level of ATP generated is determined by photons produced by luciferase. The nucleic acids to be sequenced can be attached to elements in an array and the array can be imaged to capture the chemiluminescent signals that are produced as a result of the incorporation of nucleotides into the array elements. The image can be obtained after processing the array using a specific type of nucleotide (for example, A, T, C or G). The images obtained after adding each type of nucleotide will differ depending on what elements are found in the array. These differences in the image reflect the different contents of the sequence of array elements. However, the relative locations of each element will remain the same in the images. These images may be stored, processed and analyzed using the methods described herein. For example, images obtained after processing an array using each different type of nucleotides can be processed in the same manner as shown herein as an example for images obtained from different detection channels for reversible terminator sequencing methods].

[115] According to another exemplary type of sequencing by synthesis, the sequencing cycle is carried out by sequentially adding nucleotides containing reversible terminators, for example, a cleavable tag or a photobleaching dye tag, as described, for example, in WO 04/018497 and US patent No. 7,057,026, the disclosure of which is incorporated herein by reference. This approach is commercialized by Solexa (now Illumina Inc.) and described in WO 91/06678 and WO 07/123,774, each of which is incorporated herein by reference. The presence of fluorescently labeled terminators, for which termination can be reversed and the fluorescent tag cleaved, facilitates efficient cycle reversible termination (CRT) sequencing. Additionally, polymerases can be designed to efficiently incorporate and propagate from these altered nucleotides.

[116] In embodiments of reversible terminator sequencing, the tags may not substantially prevent extension under the conditions of the synthesis sequencing reaction. However, detection marks may be removable, for example by cleavage or decomposition. Images can be captured after incorporating tags into the nucleic acid array elements. In specific embodiments, each cycle involves simultaneous delivery of four different types of nucleotides to the array, with each type of nucleotide having a spectrally distinct label. Four images can then be acquired, each using a detection channel that is selective for one of four different marks. Alternatively, different types of nucleotides can be added sequentially and an image of the array can be obtained between addition steps. In such embodiments, each image will represent nucleic acid elements that include a specific type of nucleotide. Different elements will be present or absent in different images due to the different sequence content of each element. However, the relative positions of the elements will remain unchanged in the images. Images obtained using such reversible terminator sequencing methods can be stored, processed and analyzed as described herein. After the image capture step, the tags can be removed and the reversible terminator components can be removed to allow subsequent rounds of nucleotide addition and detection. Removing tags after they are detected in a particular cycle and before the subsequent cycle may provide the benefit of reducing background signal and cross-talk between cycles.

[117] In certain embodiments, some or all of the nucleotide monomers may include reversible terminators. In such embodiments, the reversible terminators/cleavable fluorophores may include a fluorine linked to the ribose moiety by a 3' ester linkage (Metzker, Genome Res. 15:1767-1776 (2005), which is incorporated herein by reference). Other approaches separate the chemical analysis of the terminator from the cleavage of the fluorescent tag (Ruparel et al., Proc Natl Acad Sci USA 102: 5932-7 (2005), which is incorporated herein by reference in its entirety). Ruparel et al. described the development of reversible terminators using a small 3'-allyl group to block extension, but which could be easily unblocked by a short treatment using a palladium catalyst. The fluorophore was attached to the base by a photocleavable linker, which is easily cleaved by 30 s of exposure to long-wave ultraviolet light. Thus, either disulfide reduction or photocleavage can be used as cleavable binding. Another approach to reversible termination is to use natural termination, which occurs after placing a bulk dye on a deoxyribonucleoside triphosphate (dNTP). The presence of a charged bulk dye on the deoxyribonucleoside triphosphate can act as an effective terminator due to steric mismatch and/or electrostatic interference. The presence of one turn-on event prevents further turn-ons until the dye is removed. When the dye is broken down, fluoride is removed and termination is effectively reversed. Examples of altered nucleotides are also described in US Patent No. 7,427,673 and US Patent No. 7,057,026, the disclosures of which are incorporated herein by reference in their entirety.

[118] Additional exemplary sequencing-by-synthesis systems and methods that can be used with the methods and systems described herein are described in US Patent Application Publication No. 2007/0166705, US Patent Application Publication No. 2006/0188901, US Patent No. 7,057,026 , US Patent Application Publication No. 2006/0240439, US Patent Application Publication No. 2006/0281109, PCT Publication No. WO 05/065814, US Patent Application Publication No. 2005/0100900, PCT Publication No. WO 06/064199, PCT Publication No. WO 07/010,251, US Patent Application Publication No. 2012/0270305, and US Patent Application Publication No. 2013/0260372, the disclosures of which are incorporated herein by reference in their entirety.

[119] In some embodiments, detection of four different nucleotides using fewer than four different labels can be used. For example, sequencing by synthesis can be performed using methods and systems described in US Patent Application Publication No. 2013/0079232. As a first example, a pair of nucleotide types may be detected at the same wavelength, but are differentiated based on a difference in intensity for one member of the pair compared to the other, or based on a change in one member of the pair (e.g., by chemical change, photochemical change or physical change) that causes the observed signal to appear or disappear relative to the signal detected for the other member of the pair. As a second example, nucleotides belonging to three of the four different types may be detected under certain conditions, while nucleotides of the fourth type lack a label that is detectable under those conditions or is minimally detectable under those conditions (e.g., it may minimal detection due to background fluorescence, etc.). The inclusion of the first three types of nucleotides in a nucleic acid can be determined based on the presence of their corresponding signals, and the inclusion of a fourth type of nucleotide in a nucleic acid can be determined based on the absence or minimal detection of any signal. As a third example, one type of nucleotide may include tag(s) that are detected in two different channels, whereas other types of nucleotides are detected in at most one of the channels. The above three configuration examples are not considered mutually exclusive and can be used in various combinations. An embodiment that combines all three examples is a fluorescence-based synthesis sequencing method that uses a first type of nucleotide that is detected in a first channel (e.g., deoxyadenosine triphosphate (dATP) with a tag that is detected in the first channel upon excitation using a first excitation wavelength), a second type of nucleotide that is detected in a second channel (e.g., deoxycytidine triphosphate (dCTP) with a label that is detected in a second channel when excited using a second excitation wavelength), a third type of nucleotide that is detected as in the first and second channels (e.g., trimidine triphosphate (dTTP) with at least one label that is detectable in both channels when excited using the first and/or second excitation wavelength), and a fourth type nucleotide that lacks label and which is undetectable or minimally detectable in any channel (eg, deoxyguanosine triphosphate (dGTP) without a label).

[120] Additionally, as described in US Patent Application Publication No. 2013/0079232 incorporated herein, sequencing data can be generated using a single channel. In such so-called single-dye sequencing approaches, the first type of nucleotide is labeled, but said label is removed after the first image is generated, and the second type of nucleotide is labeled only after the first image is generated. The third type nucleotide retains its label in the first and second images, and the fourth type nucleotide does not contain a label in both images.

[121] In some embodiments, sequencing using ligation techniques may be used. Such methods use DNA ligase to incorporate oligonucleotides and identify the inclusion of said oligonucleotides. Oligonucleotides typically have different labels that map to the identity of the particular nucleotide in the sequence to which the oligonucleotides hybridize. As with other synthetic sequencing methods, images can be obtained after processing an array of nucleic acid elements with labeled sequencing reagents. Each image will represent nucleic acid elements that include specific types of nucleotide tags. Different elements will be present or absent in different images due to the different sequence contents of each element, but the relative positions of the elements will remain the same across the images. Images obtained using ligation-based sequencing methods can be stored, processed and analyzed as described herein. Exemplary sequencing-by-synthesis systems and methods that may be used with the methods and systems described herein are described in U.S. Patent No. 6,969,488, U.S. Patent No. 6,172,218, and U.S. Patent No. 6,306,597, the disclosure of which is incorporated herein by reference in its entirety.

[122] In some embodiments, nanopore sequencing may be used (Deamer, D. W. & Akeson, M. “Nanopores and nucleic acids: prospects for ultrarapid sequencing.” Trends Biotechnol. 18, 147-151 (2000); Deamer, D. and D. Branton, “Characterization of nucleic acids by nanopore analysis,” Acc. Chem. Res. 35:817–825 (2002); Li, J., M. Gershow, D. Stein, E. Brandin, and J. A. Golovchenko, "DNA molecules and configurations in a solid-state nanopore microscope" Nat. Mater. 2:611-615 (2003), the disclosure of which is incorporated herein by reference in its entirety. In such embodiments, the target nucleic acid passes through the nanopores. The nanopore may be a synthetic pore or a biological membrane protein such as α-hemolysin. As the target nucleic acid passes through the nanopore, each base pair can be identified by measuring fluctuations in the electrical conductivity of the pore. (US Patent No. 7,001,792; Soni, G.V. and Meller, “A. Progress toward ultrafast DNA sequencing using solid-state nanopores.” Clin. Chem. 53, 1996-2001 (2007); Healy, K. “Nanopore-based single- molecule DNA analysis." Nanomed. 2, 459-481 (2007); Cockroft, S. L., Chu, J., Amorin, M. and Ghadiri, M. R. "A single-molecule nanopore device detects DNA polymerase activity with single-nucleotide resolution" J. Am. Chem. Soc. 130, 818-820 (2008), the disclosure of which is incorporated herein by reference in its entirety). Data obtained from nanopore sequencing can be stored, processed and analyzed as described herein. In particular, the data may be processed as an image in accordance with typical optical and other image processing that is described herein.

[123] In some embodiments, methods may be used that include real-time monitoring of DNA polymerase activity. Nucleotide inclusions can be detected through fluorescence resonance energy transfer (FRET) interactions between a fluorophore-containing polymerase and γ-phosphate-labeled nucleotides, as described, for example, in US Pat. No. 7,329,492 and US Pat. No. 7,211,414 (each of which is incorporated herein by reference ) or nucleotide inclusions can be detected using zero-mode waveguides, as described, for example, in US Patent No. 7,315,019 (which is incorporated herein by reference) and using fluorescent nucleotide analogues and engineered polymerases, as described, for example, in the patent US No. 7,405,281 and US Patent Application Publication No. 2008/0108082 (each of which is incorporated herein by reference). Illumination can be limited to a volume on the order of a zeptolitre around the surface-bound polymerase, so that the incorporation of fluorescently labeled nucleotides can be observed at low background (Levene, MJ, et al. “Zero-mode waveguides for single-molecule analysis at high concentrations.” Science 299, 682-686 (2003); Lundquist, P. M. et al. "Parallel confocal detection of single molecules in real time." Opt. Lett. 33, 1026-1028 (2008); Korlach, J. et al. "Selective aluminum passivation for targeted immobilization of single DNA polymerase molecules in zero-mode waveguide nano structures." Proc. Natl. Acad. Sci. USA 105, 1176-1181 (2008), the disclosure of which is incorporated herein by reference in its entirety). Images obtained using such methods can be stored, processed and analyzed as described herein.

[124] Some embodiments of sequencing by synthesis involve detecting the proton released upon incorporation of a nucleotide into the extension product. For example, sequencing based on the detection of released protons may use an electrical detector and associated methods that are commercially available from Ion Torrent (Guilford, CT, a subsidiary of Life Technologies), or the sequencing methods and systems described in 2009/ 0026082 A1 (USA); 2009/0127589 A1 (USA); 2010/0137143 A1 (USA) or 2010/0282617 A1 (USA), each of which is incorporated herein by reference. The methods described herein for amplifying target nucleic acids using kinetic exclusion can be readily applied to substrates used for proton detection. In particular, the methods described herein can be used to generate clonal amplicon populations that are used for proton detection.

[125] The above synthetic sequencing methods may preferably be carried out in multiplex formats such that multiple different target nucleic acids are used simultaneously. In specific embodiments, different target nucleic acids may be processed in a common reaction vessel or on the surface of a specific support. This allows for convenient delivery of sequencing reagents, removal of unreacted reagents, and detection of inclusion events in a multiplexed manner. In embodiments using surface-bound target nucleic acids, said acids may be in an array format. In an array format, target nucleic acids can typically be bound to a surface in a spatially distinct manner. Target nucleic acids can be linked by direct covalent attachment, attachment to a microcarrier or other particle, or by binding to a polymerase or other molecule that is attached to a surface. The array may include one copy of the target nucleic acid at each site (also called an element), or multiple copies of the same sequence may be present at each site or element. Multiple copies can be obtained using amplification techniques such as bridge amplification or emulsion polymerase chain reaction.

[126] The methods described herein may employ arrays having elements with any of a variety of densities, including, for example, at least about 10 elements/cm2, 100 elements/cm2, 500 elements/cm2, 1000 elements/cm2. cm2, 5000 elements/cm2, 10,000 elements/cm2, 50,000 elements/cm2, 100,000 elements/cm2, 1,000,000 elements/cm2, 5,000,000 elements/cm2 or higher.

[127] The methods described herein can provide rapid and efficient simultaneous detection of multiple target nucleic acids. Accordingly, the present disclosure provides integrated systems capable of preparing and detecting nucleic acids using methods known in the art, such as the methods exemplified above. Thus, the integrated system of the present disclosure may comprise fluid components capable of delivering amplification reagents and/or sequencing reagents to one or more immobilized DNA fragments, the system including components such as pumps, valves, reservoirs, fluid lines, etc. n. The flow cell can be installed and/or used in an integrated system for detecting target nucleic acids. Exemplary flow cells are described, for example, in 2010/0111768 A1 (US) and No. 13/273,666 (US), each of which is incorporated herein by reference. As shown for flow cells, one or more fluid components of the integrated system can be used for the amplification method and for the detection method. Considering the embodiment of nucleic acid sequencing as an example, one or more liquid components of an integrated system can be used for the amplification method described herein and for delivering sequencing reagents in a sequencing method, such as the method exemplified above. Alternatively, the integrated system may comprise separate fluid systems for carrying out the amplification methods and for carrying out the detection methods. Examples of integrated sequencing systems that are capable of generating amplified nucleic acids as well as determining the sequence of nucleic acids include, but are not limited to, the MiSeq™ platform (Illumina, Inc., San Diego, Calif.) and the devices described in US Pat. No. 13/273,666 , which is incorporated herein by reference.

[128] In some embodiments of the methods described herein, the mapped sequence tags comprise sequence reads of about 20 base pairs, about 25 base pairs, about 30 base pairs, about 35 base pairs, about 40 base pairs, approximately 45 base pairs, approximately 50 base pairs, approximately 55 base pairs, approximately 60 base pairs, approximately 65 base pairs, approximately 70 base pairs, approximately 75 base pairs, approximately 80 base pairs, approximately 85 base pairs, approximately 90 base pairs, approximately 95 base pairs, approximately 100 base pairs, approximately 110 base pairs, approximately 120 base pairs, approximately 130 base pairs, approximately 140 base pairs, approximately 150 base pairs bases, about 200 base pairs, about 250 base pairs, about 300 base pairs, about 350 base pairs, about 400 base pairs, about 450 base pairs, or about 500 base pairs. In some cases, single-end reads over 500 base pairs are used for reads over approximately 1000 base pairs when generating a paired-end read. Sequence tag mapping is achieved by comparing the tag sequence to a reference sequence to determine the chromosomal origin of the nucleic acid molecule being sequenced, without requiring specific genetic sequence information. A small degree of mismatch (0-2 mismatches per sequence tag) may be allowed to account for minor polymorphisms that may exist between the reference genome and the genomes in the mixed sample.

SYSTEMS AND DEVICES FOR ANALYSIS OF REAL-TIME SEQUENCING DATA

[129] Analysis of sequencing data is typically performed using various algorithms and computer-executed programs. Thus, in some embodiments, processes are used that include storing data on one or more computer systems or other processing systems or transmitting data through one or more computer systems or other processing systems. The embodiments disclosed herein also relate to an apparatus for performing the above operations. Said device may be particularly configured to perform the required tasks, or may be a general purpose computer (or group of computers) selectively activated or reconfigurable by a computer program and/or data structure stored in the computer. In some embodiments, a group of processors performs some or all of the described analytical operations jointly (eg, via a network or cloud computing) and/or in parallel. The processor or group of processors for performing the methods described herein may be of various types, including microcontrollers and microprocessors, for example, programmable devices (for example, complex programmable logic integrated circuit (CPLD) and field programmable logic integrated circuit (FPGA)), and non-programmable devices, such as special-purpose integrated circuit (ASIC) gate arrays or general-purpose microprocessors.

[130] In addition, some embodiments relate to tangible and/or computer readable media intended for non-transitory storage of information, or to computer program products that include program instructions and/or data (including data structures) to perform various operations implemented with using a computer. Examples of computer-readable media include, but are not limited to, semiconductor storage devices, magnetic media such as disk drives, magnetic tape, optical media such as compact discs, magneto-optical media, and hardware devices that, in particular, are configured to store and executing software commands such as read only memory (ROM) and random access memory (RAM). The end user may directly or indirectly control the computer-readable medium. Examples of directly managed media include media located on user equipment and/or media that is not shared. Examples of indirectly controlled media include media that is indirectly accessible to a user through an external network and/or through a service that provides shared resources, such as a “cloud computing environment.” Examples of program instructions include both machine code, such as that produced by a compiler, and files containing higher level code that can be executed by a computer using an interpreter.

[131] In various embodiments, the data or information used in the disclosed methods and apparatus is provided in electronic format. Such data or information may include reads derived from a nucleic acid sample, numbers or densities of tags that match specific regions of the reference sequence (e.g., that match a chromosome or chromosome segment), separating distances between adjacent reads or fragments, distributions of such separating distances , diagnoses, etc. For the purposes of this document, data or other information provided in electronic format is available for storage on a machine and transmission between machines. Typically, data in electronic format is obtained in digital form and can be stored in the form of bits and/or bytes in various data structures, lists, databases, etc. Data can be recorded electronically, optically, etc.

[132] In one embodiment, a computer program product is provided for determining phasing and phasing advance coefficients, as well as phasing corrected magnitude values and associated base discriminations. The computer product may contain instructions to perform any one or more of the above-described methods for phasing and base recognition. As explained above, a computer product may include a non-transitory and/or tangible computer-readable medium having computer-executable or compiled logic (e.g., instructions) stored therein to enable a processor to align reads, identify fragments, and/or aligned islands. reads, identify alleles, including alleles resulting from insertion-deletion mutations, heterozygous polymorphisms, phase fragments of chromosomes, as well as chromosomes and genomes of haplotypes. In one example, a computer product includes (1) a computer-readable medium on which a computer-executable or compiled logic procedure (e.g., instructions) is stored to enable a processor to perform phasing correction on quantity data (e.g., color intensity data from two or more channels) on nucleic acid samples; (2) a computer-operated logic procedure for performing base recognitions on the nucleic acid samples; and (3) an inference procedure for generating output data characterizing the nucleic acid samples.

[133] It should be understood that it would be impractical, or even in most cases impossible, for a person to perform computational operations according to the methods disclosed herein. For example, generating phasing coefficients for even a single segment during a single base recognition cycle may require years of work without the use of a computing device. Of course, the problem is compounded by the fact that reliable NGS (next generation sequencing) sequencing typically requires phasing correction and base discrimination for at least thousands or even millions of reads.

[134] The methods disclosed herein can be performed using a nucleic acid sample sequencing system. The system may comprise: (a) a sequencer for receiving nucleic acids from a test sample, providing sequence information about the nucleic acid from the sample; (b) processor; and (c) one or more computer-readable media storing instructions for execution on a processor to evaluate data from the sequencer. The computer-readable medium may also store partially phase-corrected quantity data from clusters in the flow cell.

[135] In some embodiments, instructions to perform the methods are received from a computer-readable medium on which computer-readable instructions for implementing a method for determining a sequence phase are stored. Thus, one embodiment provides that the computer program product uses one or more machine-readable non-transitory storage media on which computer-executable instructions are stored that, when executed by one or more processors of a computer system, the computer system implements a method for sequencing a DNA sample. The method includes: (a) obtaining data representing an image (eg, the image itself) of a substrate containing a plurality of sites on which nucleic acid bases are read; (b) obtaining color values (or other values representing individual bases/nucleotides) for a plurality of regions from the substrate image; (c) storing color values in a processor buffer; (d) retrieving partial phasing corrected color values for a plurality of sites for a base recognition cycle, wherein the partial phasing correction color values were stored in the sequencer memory during the immediately preceding base recognition cycle; (e) determining the advanced phasing correction from (i) the partially phasing corrected color values stored during the immediately preceding base recognition cycle and (ii) the color values stored in the processor buffer; and (f) determining corrected color values from (i) color values in the processor buffer, (ii) partially phasing-corrected values stored during the immediately preceding cycle, and (iii) phasing-advance correction.

[136] The sequence or other data may be entered into a computer or stored on a machine-readable medium, directly or indirectly. In various embodiments, the computer system is embedded in or directly coupled to a sequencing device that reads and/or analyzes nucleic acid sequences from samples. Sequences or other information from such instruments are transmitted to a computer system (or simply to embedded processing hardware) via a data interface. In addition, the storage device may store reads, base recognition quality information, phasing coefficient information, etc. The storage device may also store various routines and/or programs for analyzing and reporting sequence data. Such programs/procedures may include programs to perform statistical analysis, etc.

[137] In one example, the user provides the sample to the sequencing device. A sequencing device connected to a computer collects and/or analyzes data. Software installed on a computer allows data to be collected and/or analyzed. The data may be stored, displayed (using a monitor or other similar device) and/or sent to another location. The computer may be connected to the Internet, which is used to transmit data to a mobile device used by a remote user (for example, a doctor, scientist, or analyst). It is understood that the data may be stored and/or analyzed prior to transmission. In some embodiments, the raw data is collected and sent to a remote user or device that will analyze and/or store said data. For example, reads can be transmitted as they are generated or shortly thereafter and reconciled, while others can be analyzed remotely. The transmission may take place over the Internet, but may also be via satellite or other connection. Alternatively, the data may be stored on a machine-readable medium and the medium may be sent to an end user (eg, by mail). The remote user may be in the same or a different geographic location, including, without limitation, a building, city, state, country or continent.

[138] In some embodiments, the methods also include collecting data relating to multiple polynucleotide sequences (eg, reads) and sending said data to a computer or other computing system. For example, the computer may be connected to laboratory equipment, such as a sample collection device, a polynucleotide amplification device, or a nucleotide sequencing device. Collected or stored data may be transferred from a computer to a remote location, such as a local area network or a wide area network such as the Internet. At a remote location, various operations can be performed on the transmitted data.

[139] In some embodiments of any of the systems presented herein, the sequencer is configured to perform next generation sequencing (NGS). In some embodiments, the sequencer is configured to perform massively parallel sequencing using sequencing by synthesis using reversible dye terminators. In other embodiments, the sequencer is configured to perform single molecule sequencing.

CONCLUSION

[140] The present disclosure may be embodied in other specific forms without departing from its spirit or essential characteristics. The described embodiments are to be considered in all respects only as illustrative and not limiting. Accordingly, the scope of the disclosure is determined by the appended claims and not by the above description. All changes that come within the scope and scope of the claims must be included within its scope.

Claims (47)

1. A method of determining corrected color values from image data obtained during a base recognition cycle using a nucleic acid sequencer comprising an image acquisition system, one or more processors and a storage device, comprising: (a) obtaining an image of a substrate containing a plurality of regions on which nucleic acid bases are read, these regions having colors representing types of nucleic acid bases; (b) measuring color values of a plurality of regions from said substrate image; (c) storing the color values in a processor buffer of one or more sequencer processors; (d) retrieving phasing-corrected color values of a plurality of regions where said phasing-corrected color values were stored in the sequencer memory during the immediately preceding base recognition cycle; And (e) determination of corrected color values by color values in the processor buffer and phasing-corrected color values stored from the immediately preceding cycle. 2. The method of claim 1, wherein the phasing advance correction includes a weight, and wherein determining the corrected color values includes multiplying the weight by the color values of a plurality of regions measured from the substrate image. 3. The method according to claim 1, additionally including: determining a phasing correction for the immediately following base recognition cycle; obtaining phasing-corrected color values for the immediately following base discrimination cycle by applying phasing correction to the color values of multiple sites stored in sequencer memory by summing color values with phasing correction of multiple sections and color values of a plurality of regions from the substrate image measured in step (b); And storing phasing-corrected color values for the immediate next base recognition cycle in the sequencer memory. 4. The method according to claim 1, in which the nucleic acid sequencer synthesizes nucleic acids at multiple sites. 5. The method according to claim 1, in which the color values are determined using two channels of the sequencer. 6. The method according to claim 1, in which the color values are obtained from four channels of the sequencer. 7. The method of claim 1, wherein the substrate comprises a flow cell, wherein the flow cell is logically divided into segments, wherein each segment represents a region of the flow cell containing a subset of regions where said subset is captured in a single image from the imaging system, and wherein the method also includes, prior to step (a), placing the reagents in a flow cell and allowing the reagents to interact with the patches, causing the patches to have colors representing the types of nucleic acid bases during the base recognition cycle. 8. The method according to claim 7, which also includes after operation (e): placing fresh reagents in a flow cell and allowing the fresh reagents to interact with the patches, causing the patches to have colors representing the types of nucleic bases; And repeating operations (a)-(e) for the next base recognition cycle. 9. The method of claim 8, which further includes creating a first processor thread to perform operations (a)-(e) for the base recognition cycle and creating a second processor thread to perform operations (a)-(e) for the next base recognition cycle. 10. Nucleic acid sequencer containing: image acquisition system; Memory device; And one or more processors configured to: (a) obtaining data representing an image of a substrate containing a plurality of regions on which nucleic acid bases are read, said regions having colors representing types of nucleic acid bases; (b) obtaining color values of a plurality of regions from the substrate image; (c) storing color values in a processor buffer; (d) retrieving the phasing-corrected color values of the plurality of regions for a base recognition cycle, the phasing-corrected color values having been stored in a storage device during the immediately preceding base recognition cycle; And (e) determination of corrected color values by color values in the processor buffer and phasing-corrected color values stored from the immediately preceding cycle. 11. The nucleic acid sequencer of claim 10, wherein the storage device is divided into a plurality of segment buffers, each of which is configured to store data representing one segment image on the substrate. 12. The nucleic acid sequencer of claim 10, wherein the storage capacity is 512 gigabytes or less. 13. The nucleic acid sequencer of claim 10, wherein the phasing advance correction includes a weight, and wherein one or more processors are configured or configured to determine the corrected color values by multiplying the weight by the color values of the plurality of regions measured from the substrate image. 14. The nucleic acid sequencer of claim 10, wherein the one or more processors are also configured or configured to: determining phasing correction for the immediate next base recognition cycle; obtaining phasing-corrected color values for the immediately following base discrimination cycle by applying phasing correction to the color values of the plurality of regions stored in the memory device, and storing the phasing-corrected color values for the immediately following base recognition cycle in the memory device. 15. The nucleic acid sequencer of claim 14, wherein one or more processors are configured or configured to obtain phasing-corrected color values for the immediately following base cycle by summing color values with phasing correction of multiple sections and color values of a plurality of regions from the substrate image measured in step (b). 16. The nucleic acid sequencer of claim 14, wherein the one or more processors are configured or configured to store phase-corrected color values for the immediately following base recognition cycle by storing the phase-corrected color values in memory segment buffers. 17. A nucleic acid sequencer according to claim 10, further comprising a system for synthesizing nucleic acids at multiple sites. 18. The nucleic acid sequencer of claim 10, wherein the one or more processors are configured or configured to receive color values from the two channels.