WO2025199147A1 - Methods and systems for massively scalable drug discovery through chemical imaging and machine learning - Google Patents

Abstract

Provided herein are methods of using a machine learning model to analyze a combination of imaging data and sequencing data to characterize cellular phenotypes and evaluate perturbation effects of drugs and genetic perturbation effects.

Description

Attorney Docket No.: 51812-006WO2 PATENT METHODS AND SYSTEMS FOR MASSIVELY SCALABLE DRUG DISCOVERY THROUGH CHEMICAL IMAGING AND MACHINE LEARNING Field of the Invention The invention relates to methods and systems for drug discovery using vibrational microscopy in combination with cell expression profiling. Cross-Reference to Related Applications This application claims benefit of U.S. Provisional Application No.63/566,426 filed March 18, 2024, the content of which is incorporated by reference. Background of the Invention Drug discovery processes are notoriously slow, costly, and inefficient. While phenotypic screens offer promise in expediting drug discovery, existing methods encounter numerous challenges. An ideal method must meet specific criteria: (1) ensuring a robust signal-to-noise ratio at the single-cell level, (2) providing high throughput capabilities, (3) delivering comprehensive content, (4) maintaining non- invasiveness, (5) being cost-effective and user-friendly, (6) minimizing batch effects, and (7) compatible with different perturbations such as small molecule, genetic perturbations in arrayed or pooled format. Fluorescent image-based profiling methods, such as cell painting and fluorescent reporters, offer detailed morphological insights but are susceptible to batch and layout effects, limited signal-to-noise ratio, significantly influenced by cell density. Mass spectrometry, although offering diverse metabolic or antigenic features, is both destructive and expensive. Single-cell RNA sequencing can unveil drug- induced molecular changes but is costly, operationally complex, and difficult to scale up. Vibrational spectroscopy and chemical imaging techniques hold promise in providing a comprehensive biochemical profile of cell biochemical phenotypes and functions, utilizing the vibrational spectrum of molecules within individual cells. These techniques, being non-destructive, offer high- dimensional biochemical features (~200X more dimensions compared to fluorescent imaging). Nonetheless, their potential for drug discovery remains largely unexplored. New techniques are needed for efficient and effective drug discovery processes. Summary of the Invention In one aspect, the invention features a method of analyzing a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data. In one aspect, the invention features a method of identifying a biological state in a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ Attorney Docket No.: 51812-006WO2 PATENT sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates the biological state. In one aspect, the invention features a method of determining the efficacy of an agent in effecting a change in a biological sample comprising: (a) obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample after administration of the agent wherein the sequencing data are obtained using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates that the agent is effective in effecting a change in the biological sample. In one aspect, the invention features a method of analyzing a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to the one or more images and the sequencing data to analyze the one or more images. In one aspect, the invention features a method of identifying a biological state in a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to determine the biological state. In one aspect, the invention features a method of determining the efficacy of an agent in effecting a change in a biological sample comprising: (a) obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to determine that the agent is effective in effecting a change in the biological sample. In one aspect, the invention features a computer whose input data is: (a) images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; or (b) images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing, wherein the computer is programmed with a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing. In one aspect, the invention features a system comprising: (a) an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) a single-cell profiling component configured for single-cell sequencing and/or in situ sequencing; and (c) a computer configured to receive images from the imaging component and sequencing data from the single-cell profiling component, wherein the imaging component and the single- Attorney Docket No.: 51812-006WO2 PATENT cell profiling component are each operatively coupled to the computer, wherein the computer is programed with a machine learning model whose input data is the images and the sequencing data, and wherein the machine learning model correlates the images with the sequencing data. In one aspect, the invention features a system comprising: (a) an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) a computer configured to receive images from the imaging component, wherein the imaging component is operatively coupled to the computer, and wherein the computer comprises a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing. In some embodiments, the machine learning model was developed using a combination of: (a) an imaging technique comprising infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) a sequencing technique comprising single-cell sequencing and/or in situ sequencing. In some embodiments, the machine learning model is a supervised learning model, an unsupervised learning model, a self-supervised learning model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model. In some embodiments, each of the one or more images comprises a collection of pixels, wherein each pixel comprises an infrared spectrum, a Raman spectrum, or a combination thereof. In some embodiments, the method further comprises determining a chemical composition of the biological sample at each pixel. In some embodiments, the one or more images are 3D images. In some embodiments, the 3D images comprise optical sections, and wherein the optical sections comprise a thickness of 5 µm to 500 µm. In some embodiments, the infrared microscopy produces a spectrum from 400 to 4000 cm^-1 and/or the Raman microscopy produces a spectrum from 500 to 3500 cm^-1. In some embodiments, the spectrum is from about 900 to about 1800 cm^-1. In some embodiments, the agent is a small molecule, a nucleic acid, a protein, a gene editing agent, or a cell. In some embodiments, the nucleic acid is an RNAi, an shRNA, an mRNA, or an antisense oligonucleotide. In some embodiments, the protein is a transcription factor, an antibody, a peptide, a cytokine, a hormone, an enzyme, an antibody drug conjugate, or a fusion protein. In some embodiments, the gene editing agent is clustered regularly interspaced short palindromic repeats (CRISPR), a zinc finger nuclease, a base editor, a prime editor, or transcription activator-like effectors (TALEN). In some embodiments, the cell is a chimeric antigen receptor T (CAR-T) cell. In some embodiments, the biological sample comprises a cell, a tissue, and/or a cellular component. In some embodiments, the biological sample was obtained from a human. In some embodiments, the biological sample is obtained from a subject with a disorder. In some embodiments, the disorder is an infectious disease (e.g., resulting from a bacteria, virus, fungus, or parasite), a cardiovascular disease, a respiratory disease, a metabolic disorder, an endocrine disorder, a neurological disorder, a mental disorder, an autoimmune disease, a gastrointestinal disease, a rare disease. In some embodiments, the biological sample comprises human embryonic stem cells. Attorney Docket No.: 51812-006WO2 PATENT In some embodiments, the effected change comprises treating a disorder, altering a gene or protein expression profile, inducing cell differentiation, inducing cell death, or a combination thereof. Other features and advantages of the invention will be apparent from the following detailed description and figures, and from the claims. Brief Description of the Drawings The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application with color drawings will be provided by the Office upon request and payment of the necessary fee. FIG.1 is a graphic demonstrating methods and systems for massively scalable drug discovery through chemical imaging and machine learning. FIG.2 is a graphic representing the experimental design in Example 2. FIG.3 is a plot showing hierarchical clustering of mean spectra of each sample of human embryonic stem cells (hESCs) treated with transcription factors (TFs; TFs indicated on x-axis were coded according to Table 1). Clusters of cells with similar results are shown. Definitions It is to be understood that aspects and embodiments of the invention described herein include “comprising,” “consisting,” and “consisting essentially of” aspects and embodiments. As used herein, the singular form “a,” “an,” and “the” includes plural references unless indicated otherwise. The term “about” as used herein refers to the usual error range for the respective value readily known to the skilled person in this technical field. Reference to “about” a value or parameter herein includes (and describes) embodiments that are directed to that value or parameter per se. In some instances, reference to “about” a value or parameter herein indicates the value or parameter ± 10%. The term “computer” as used herein may refer to device embodied in any of a number of forms, such as a rack-mounted computer, a desktop computer, a laptop computer, or a tablet computer, as non- limiting examples. Additionally, a computer may be embedded in a device not generally regarded as a computer but with suitable processing capabilities, including a Personal Digital Assistant (PDA), a smartphone or any other suitable portable or fixed electronic device. A computer may have one or more communication devices, which may be used to interconnect the computer to one or more other devices and/or systems, such as, for example, one or more networks in any suitable form, including a local area network or a wide area network, such as an enterprise network, and intelligent network (IN) or the Internet. Such networks may be based on any suitable technology and may operate according to any suitable protocol and may include wireless networks or wired networks. A computer may have one or more input devices and/or one or more output devices. These devices may be used, among other things, to present a user interface. Examples of output devices that may be used to provide a user interface include printers or display screens for visual presentation of output and speakers or other sound generating devices for audible presentation of output. Examples of input devices that may be used for a user interface include keyboards, and pointing devices, such as mice, touch pads, and digitizing tablets. As another example, a computer may receive input information through speech recognition or in other Attorney Docket No.: 51812-006WO2 PATENT audible formats. Computer-executable instructions may be in many forms, such as program modules, executed by one or more computers or other devices. Generally, program modules include routines, programs, objects, components, data structures, etc. that perform particular tasks or implement particular abstract data types. The functionality of the program modules may be combined or distributed as desired in various embodiments. Databases, if employed in the methods or devices or systems herein, may include computer readable memory (also referred to as “memory”). For example, data storage space 3memlN may be and/or include computer readable memory, used to store data as described in the disclosure. Memory may be embodied by suitable hardware, including but not limited to the following: hard disk drives, serial advanced technology attachment (SATA) hard drives, SATA solid state drives (SSDs), non-volatile memory express (NVMe) SSDs, tape drives. The terms “program,” “app,” and “software” are used herein in a generic sense to refer to any type of computer code or set of computer-executable instructions that may be employed to program a computer or other processor to implement various embodiments described herein. Additionally, it should be appreciated that, according to one aspect, one or more computer programs that when executed perform methods of this application need not reside on a single computer or processor but may be distributed in a modular fashion among a number of different computers or processors to implement various embodiments of this application. As used herein, the term “biological sample” refers to a subset of its tissues, cells or component parts (e.g. body fluids, including but not limited to peripheral blood, serum, plasma, ascites, urine, cerebrospinal fluid (CSF), sputum, saliva, bone marrow, synovial fluid, aqueous humor, amniotic fluid, cerumen, breast milk, broncheoalveolar lavage fluid, semen, prostatic fluid, cowper's fluid or pre- ejaculatory fluid, sweat, fecal matter, hair, tears, cyst fluid, pleural and peritoneal fluid, pericardial fluid, lymph, chyme, chyle, bile, interstitial fluid, menses, pus, sebum, vomit, vaginal secretions, mucosal secretion, stool water, pancreatic juice, lavage fluids from sinus cavities, bronchopulmonary aspirates, blastocyl cavity fluid, and umbilical cord blood). A sample further may include a homogenate, lysate or extract prepared from a whole organism or a subset of its tissues, cells or component parts, or a fraction or portion thereof, including but not limited to, for example, plasma, serum, spinal fluid, lymph fluid, the external sections of the skin, respiratory, intestinal, and genitourinary tracts, tears, saliva, milk, blood cells, tumors, organs. A sample further refers to a medium, such as a nutrient broth or gel, which may contain cellular components, such as proteins or nucleic acid molecule. The term “peak area” refers to the area under the curve of a peak of a spectrum (e.g., an IR or Raman spectrum). Peak area may be calculated by determining the area under the curve between an upper limit and lower limit of the spectrum (e.g., between an upper vibrational frequency limit and a lower vibrational frequency limit). Detailed Description Described herein are methods of analyzing a biological sample. The methods include obtaining one or more infrared images of the biological sample using infrared microscopy and/or one or more Raman images of the biological sample using Raman microscopy; and applying, by a computer, a machine learning model to analyze the one or more infrared images and/or the one or more Raman images. Attorney Docket No.: 51812-006WO2 PATENT Also described herein are methods of analyzing a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data. Also described herein are methods of identifying a biological state in a biological sample. The methods include obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ sequencing; and applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates the biological state. Also described herein are methods of determining the efficacy of an agent in effecting a change in a biological sample. The methods include obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; obtaining sequencing data of the biological sample after administration of the agent wherein the sequencing data are obtained using single-cell sequencing and/or in situ sequencing; and applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates that the agent is effective in effecting a change in the biological sample. Also described herein are methods of analyzing a biological sample. The methods include obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and applying, by a computer, a machine learning model to the one or more images and the sequencing data to analyze the one or more images. Also described herein are methods of identifying a biological state in a biological sample. The methods include obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and applying, by a computer, a machine learning model to determine the biological state. Also described herein are methods of determining the efficacy of an agent in effecting a change in a biological sample. The methods include obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and applying, by a computer, a machine learning model to determine that the agent is effective in effecting a change in the biological sample. Described herein is a computer (e.g., a computer which may be used with any of the methods described herein) whose input data is: images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; or images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a Attorney Docket No.: 51812-006WO2 PATENT combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing, wherein the computer is programmed with a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing. Any of the methods described herein may be performed, e.g., using a system that includes, e.g., an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; a single-cell profiling component configured for single-cell sequencing and/or in situ sequencing; and a computer configured to receive images from the imaging component and sequencing data from the single-cell profiling component, wherein the imaging component and the single-cell profiling component are each operatively coupled to the computer, wherein the computer is programed with a machine learning model whose input data is the images and the sequencing data, and wherein the machine learning model correlates the images with the sequencing data. Any of the methods described herein may be performed, e.g., using a system comprising: (a) an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) a computer configured to receive images from the imaging component, wherein the imaging component is operatively coupled to the computer, and wherein the computer comprises a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing. In any of the methods described herein, the machine learning model may have been developed using a combination of an imaging technique comprising infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and a sequencing technique comprising single-cell sequencing and/or in situ sequencing. In any of the methods described herein, the machine learning model may have been developed using a sequencing technique comprising single-cell sequencing and/or in situ sequencing. Any of the single-cell sequencing methods and/or in situ sequencing methods described herein may be used to obtain single-cell expression data from a biological sample. In some instances, the machine learning model integrates and/or links the one or more images obtained using the imaging technique with the sequencing data obtained from the sequencing technique. In a preferred example, the machine learning model integrates image data and single cell RNA-seq/in situ sequencing data to analyze and/or determine the relationship between these two data modalities. In any of the methods described herein, the machine learning model may be a supervised learning model or an unsupervised learning model. In some instances, the machine learning model may be a supervised learning model, an unsupervised learning model, a self-supervised learning model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model. Any computer described herein may implement a machine learning model including a supervised learning model or an unsupervised learning model. Any computer described herein may implement a supervised learning model, an unsupervised learning model, a self-supervised learning Attorney Docket No.: 51812-006WO2 PATENT model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model. A machine learning model described herein (e.g., a supervised learning model, an unsupervised learning model, a self-supervised learning model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model) may use data obtained from single-cell sequencing and/or in situ sequencing (e.g., single cell expression data) as the ground truth. A machine learning model described herein may be trained using data obtained from single-cell sequencing and/or in situ sequencing (e.g., single cell expression data). A machine learning model described herein may be trained using one or more images obtained using any of the imaging techniques described herein (e.g., infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof) and data obtained from single-cell sequencing and/or in situ sequencing (e.g., single cell expression data). The machine learning models described herein may be used to predict a cell state (e.g., single cell expression) of a cell within a biological sample based on one or more images obtained using any of the imaging techniques described herein (e.g., infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof). In a particular example, A machine learning model described may be trained using one or more images obtained using any of the imaging techniques described herein (e.g., infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof) and data obtained from single-cell sequencing and/or in situ sequencing (e.g., single cell expression data) and then used to predict a cell state (e.g., single cell expression) of a cell within a biological sample based on one or more images obtained using any of the imaging techniques described herein (e.g., infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof). In any of the methods described herein, each of the one or more images may include a collection of pixels, wherein each pixel includes an infrared spectrum, a Raman spectrum, or a combination thereof. Any of the methods described herein may further include determining a chemical composition of the biological sample at each pixel. In any of the methods described herein, the one or more images may be 3D images. In any of the methods described herein, the 3D images may include optical sections, and wherein the optical sections comprise a thickness of from about 5 µm to about 500 µm (e.g., about 5 µm to about 10 µm, about 5 µm to about 20 µm, about 5 µm to about 50 µm, about 10 µm to about 50 µm, about 10 µm to about 25 µm, about 50 µm to about 100 µm, about 50 µm to about 200 µm, about 100 µm to about 200 µm, about 100 µm to about 300 µm, about 200 µm to about 300 µm, about 300 µm to about 400 µm, about 400 µm to about 500 µm, about 5 µm to about 250 µm, about 250 µm to about 500 µm, about 5 µm to about 400 µm, about 5 µm to about 300 µm, about 5 µm to about 200 µm, about 5 µm to about 100 µm, about 100 µm to about 500 µm, about 200 µm to about 500 µm, about 300 µm to 500 µm, or about 400 µm to about 500 µm; e.g., about 5, 10, 15, 20, 25, 30, 40, 50, 60, 70, 80, 90, 100, 150, 200, 250, 300, 400, or 500 µm). In any of the methods described herein, the infrared microscopy may produce a spectrum from about 400 cm^-1 to about 4000 cm^-1 (e.g., about 500 cm^-1 to about 4000 cm^-1, about 800 cm^-1 to about 4000 cm^-1, about 1000 cm^-1 to about 4000 cm^-1, about 1200 cm^-1 to about 4000 cm^-1, about 1500 cm^-1 to about 4000 cm^-1, about 1800 cm^-1 to about 4000 cm^-1, about 2000 cm^-1 to about 4000 cm^-1, about Attorney Docket No.: 51812-006WO2 PATENT 2200 cm^-1 to about 4000 cm^-1, about 2500 cm^-1 to about 4000 cm^-1, about 2800 cm^-1 to about 4000 cm^-1, about 3200 cm^-1 to about 4000 cm^-1, about 3500 cm^-1 to about 4000 cm^-1, about 400 cm^-1 to about 3500 cm^-1, about 400 cm^-1 to about 3200 cm^-1, about 400 cm^-1 to about 3000 cm^-1, about 400 cm^-1 to about 2800 cm^-1, about 400 cm^-1 to about 2500 cm^-1, about 400 cm^-1 to about 2200 cm^-1, about 400 cm^-1 to about 2000 cm^-1, about 400 cm^-1 to about 1800 cm^-1, about 400 cm^-1 to about 1500 cm^-1, about 400 cm^-1 to about 1200 cm^-1, about 400 cm^-1 to about 1000 cm^-1, about 400 cm^-1 to about 800 cm^-1, about 1000 cm^-1 to about 2000 cm^-1, about 1000 cm^-1 to about 3500 cm^-1, about 2500 cm^-1 to about 3500 cm^-1, about 500 cm^-1 to about 1500 cm^-1, about 500 cm^-1 to about 3000 cm^-1, about 1000 cm^-1 to about 2000 cm^-1, about 2000 cm^-1 to about 3000 cm^-1, about 1500 cm^-1 to about 3000 cm^-1, about 800 cm^-1 to about 1000 cm^-1, about 1000 cm^-1 to about 1300 cm^-1, about 1100 cm^-1 to about 1400 cm^-1, about 1300 cm^-1 to about 1500 cm^-1, about 1400 cm^-1 to about 1600 cm^-1, about 1500 cm^-1 to about 1800 cm^-1, about 1800 cm^-1 to about 2000 cm^-1, about 2000 cm^-1 to about 2200 cm^-1, about 2000 cm^-1 to about 2300 cm^-1, about 1000 cm^-1 to about 2300 cm^-1, or about 1000 cm^-1 to about 3100 cm^-1) and/or the Raman microscopy may produce a spectrum from about 500 cm^-1 to about 3500 cm^-1 (e.g., about 800 cm^-1 to about 3500 cm^-1, about 1000 cm^-1 to about 3500 cm^-1, about 1200 cm^-1 to about 3500 cm^-1, about 1500 cm^-1 to about 3500 cm^-1, about 1800 cm^-1 to about 3500 cm^-1, about 2000 cm^-1 to about 3500 cm^-1, about 2200 cm^-1 to about 3500 cm^-1, about 2500 cm^-1 to about 3500 cm^-1, about 2800 cm^-1 to about 3500 cm^-1, about 3200 cm^-1 to about 3500 cm^-1, about 500 cm^-1 to about 3200 cm^-1, about 500 cm^-1 to about 3000 cm^-1, about 500 cm^-1 to about 2800 cm^-1, about 500 cm^-1 to about 2500 cm^-1, about 500 cm^-1 to about 2200 cm^-1, about 500 cm^-1 to about 2000 cm^-1, about 500 cm^-1 to about 1800 cm^-1, about 500 cm^-1 to about 1500 cm^-1, about 500 cm^-1 to about 1200 cm^-1, about 500 cm^-1 to about 1000 cm^-1, about 500 cm^-1 to about 800 cm^-1, about 1000 cm^-1 to about 2000 cm^-1, about 2000 cm^-1 to about 3000 cm^-1, about 1500 cm^-1 to about 3000 cm^-1, about 800 cm^-1 to about 1000 cm^-1, about 1000 cm^-1 to about 1300 cm^-1, about 1100 cm^-1 to about 1400 cm^-1, about 1300 cm^-1 to about 1500 cm^-1, about 1400 cm^-1 to about 1600 cm^-1, about 1500 cm^-1 to about 1800 cm^-1, about 1800 cm^-1 to about 2000 cm^-1, about 2000 cm^-1 to about 2200 cm^-1, about 2000 cm^-1 to about 2300 cm^-1, about 1000 cm^-1 to about 2300 cm^-1, or about 1000 cm^-1 to about 3100 cm^-1). For example, the spectrum may be from about 900 cm^-1 to about 1800 cm^-1. In any of the methods described herein including an agent, the agent may be a small molecule, a nucleic acid, a protein, a gene editing agent, or a cell. Examples of agents which are nucleic acids include an RNAi, an shRNA, an mRNA, or an antisense oligonucleotide. Examples of agents which are proteins include a transcription factor, an antibody, a peptide, a cytokine, a hormone, an enzyme, an antibody drug conjugate, or a fusion protein. Examples of agents which are gene editing agents include clustered regularly interspaced short palindromic repeats (CRISPR), a zinc finger nuclease, a base editor, a prime editor, or transcription activator-like effectors (TALEN). Examples of agents which are cells include a chimeric antigen receptor T (CAR-T) cell. In any of the methods described herein, the biological sample may include a cell, a tissue, and/or a cellular component. The biological sample may have been obtained from a human. The biological sample may have been obtained from a subject with a disorder (e.g., a cancer; e.g., a breast cancer). In some instances, the disorder may be an infectious disease (e.g., resulting from a bacteria, virus, fungus, or parasite), a cardiovascular disease, a respiratory disease, a metabolic disorder, an endocrine disorder, Attorney Docket No.: 51812-006WO2 PATENT a neurological disorder, a mental disorder, an autoimmune disease, a gastrointestinal disease, a rare disease. The biological sample may include human cancer cells. The biological sample may include human breast cancer carcinoma cells. The biological sample may include human embryonic stem cells. In such cases, the methods described herein may be used to analyze stem cell differentiation, e.g., for cell therapies. In any of the methods described herein, the effected change may include treating a disorder, altering a gene or protein expression profile, inducing cell differentiation, inducing cell death, or a combination thereof. A system described herein may include a Fourier transformed infrared (FTIR) spectrometer. A system described herein may include a Raman spectrometer. A system described herein may also include an irradiation source, a beam splitter, a moving mirror, and a fixed mirror. A system may be suitable for performing infrared microscopy, including any one or more of optical photothermal infrared (O- PTIR) microscopy, mid-infrared photothermal (MIP) microscopy, nano-FTIR spectroscopy, dual-comb photothermal microscopy, shortwave infrared photothermal (SWIP) microscopy, quantum cascade lasers (QCLS) based IR microscopy. A system described herein may be suitable for performing Raman microscopy, including stimulated Raman scattering (SRS) microscopy. To perform FTIR imaging, Agilent Cary 620 Imaging FTIR equipped with an Agilent 670-IR spectrometer and 128 × 128-pixels FPA mercury cadmium telluride (MCT) detector is used in the transmission mode. A background spectrum is collected on a clean CaF2 substrate using 128 scans at 8 cm⁻¹ spectral resolution, suggesting that the IR absorbance is measured every 4 cm^-1. Cell spectra are recorded using 64–128 scans at 8 cm⁻¹ spectral resolution. A ×25 IR objective (pixel size, 3.3 μm, 0.81 numerical aperture (NA)) is used for cell imaging. To perform spontaneous Raman imaging, spontaneous Raman imaging is performed using an upright confocal Raman microscope (Xplora, HORIBA Jobin Yvon). Cell samples are illuminated by 532 nm laser (80 mW on sample) through a 50× objective (air, NA 0.75, MPlan N, Olympus). Raman images are acquired using the point-by-point mapping mode with an acquisition time of 5s and 1× accumulation for each point measurement. The step size is set as 7 μm. The grating is set as 1200 gr/mm. Both the slit size and the hole size is set as 100 μm.500-1000 cells are imaged for each condition. To process Raman images, raw Raman spectra are first processed in the LabSpec 6 software (HORIBA). The “Despike” function is used to remove cosmic rays. The “Threshold” function is used to remove spectra with high background. The “Correction” function is used to subtract background from raw spectra by selecting a field of view without any cells. After that the spectral range from 2830 cm^-1 to 3000 cm^-1 is integrated to generate a cell image, using custom-written MATLAB scripts. The cell image is used to segment single cells and generate a cell segmentation mask with the Otsu thresholding method using the CellProfiler software. Based on the cell segmentation mask, single-cell spectra are generated by averaging all the spectra within each cell, using custom-written MATLAB scripts. For calculation of the protein-to-lipid ratio, lipid unsaturation ratio, protein synthesis rate, saturated lipid synthesis rate, and unsaturated lipid synthesis rate in any of the methods described herein: The protein-to-lipid ratio is defined as the intensity at 2930 cm^-1 divided by the intensity at 2850 cm^-1. The lipid unsaturation ratio is defined as the intensity at 3050 cm^-1 divided by the intensity at 2850 cm^-1. The protein synthesis rate is defined as the peak area of 1600-1630 cm^-1 divided by the sum of the peak area Attorney Docket No.: 51812-006WO2 PATENT of 1600-1630 cm^-1 and the peak area of 1634-1700 cm^-1. The saturated lipid synthesis rate is defined as the peak area of 2080-2130 cm^-1 divided by the peak area of 2825-2875 cm^-1. The unsaturated lipid synthesis rate is defined as the peak area of 2130-2230 cm^-1 divided by the peak area of 2825-2875 cm^-1. Any of the methods described herein may include normalization of the vibrational spectra (e.g., the IR or Raman spectra). Chemical imaging spectra typically have two major sources of technical noise that are removed with different normalization strategies. The first is a scattering effect that introduces a drift in the spectra and is removed with a baseline correction method. The second is introduced by the variable depth of the measured tissue and may be thought of as similar to the library size in transcriptomic settings. These may be removed by computing cell-specific normalization constant as the average over the whole spectra, or specific peaks. Five possible normalization methods include: “Amide I” normalizes the area of the Amide I peak (1600-1800) to 1; “Amide II” normalizes the area of the Amide II peak (1470-1570) to 1; “CH” normalizes the area of the (a)symmetric CH stretching (2815-3015) to 1; “Signal” normalizes the area of the whole spectra to 1; and “Min-Max” linearly scales and shifts the whole spectra to lie in [0, 1] In any of the methods described herein, IR datasets may be analyzed with the signal normalization and Spontaneous Raman datasets may be analyzed with the CH normalization. In any of the methods described herein, the one or more infrared images and/or the one or more Raman images may include one or more segmentation images at a spectrum from 2830 cm^-1 to 3000 cm- ¹. In any of the methods described herein, computer may apply a cell segmentation method to the one or more segmentation images to generate an output of a segmented cell. The cell segmentation method may be an Otsu thresholding method. In an exemplary implementation of the Otsu thresholding method, an image is divided into areas of foreground and background based on a moving threshold value. For every possible threshold value, the variance of the foreground and the background is calculated. Otsu thresholding finds the value of the threshold that minimizes the weighted sum of the variances of the foreground and background. For example, in some segmentation methods, the average infrared spectrum of the segmented cell may be normalized by scaling the area of the average infrared spectrum to 1; and/or the average Raman spectrum of the segmented cell may be normalized by scaling the area of a subspectrum of from 2815 cm^-1 to 3015 cm^-1 to 1. Exemplary methods for in situ sequencing and/or determining the expression profile of cells in situ that may be useful in any of the methods described herein include ISS (Ke, R. et al. In situ sequencing for RNA analysis in preserved tissue and cells. Nat. Methods 10, 857-860 (2013)), MERFISH (Chen, K. H., Boettiger, A. N., Moffitt, J. R., Wang, S. & Zhuang, X. Spatially resolved, highly multiplexed RNA profiling in single cells. Science 348, (2015)), smFISH (Codeluppi, S. et al. Spatial organization of the somatosensory cortex revealed by cyclic smFISH. biorxiv.org/lookup/doi/10.1101/276097 (2018) doi:10.1101/276097), osmFISH (Codeluppi, S. et al. Spatial organization of the somatosensory cortex revealed by osmFISH. Nat. Methods 15, 932-935 (2018)), STARMap (Wang, X. et al. Three-dimensional intact-tissue sequencing of single-cell transcriptional states. Science 361, eaat5691 (2018)), Targeted ExSeq (Alon, S. et al. Expansion Sequencing: Spatially Precise In Situ Transcriptomics in Intact Biological Systems. biorxiv.org/lookup/doi/10.1101/2020.05.13.094268 (2020) doi:10.1101/2020.05.13.094268), seqFISH+ (Eng, C.-H. L. et al. Transcriptome-scale super-resolved imaging in tissues by RNA seqFISH+. Nature (2019) doi:10.1038/s41586-019-1049-y.), Spatial Transcriptomics methods (e.g., Spatial Attorney Docket No.: 51812-006WO2 PATENT Transcriptomics (ST)) (see, e.g., Ståhl, P. L. et al. Visualization and analysis of gene expression in tissue sections by spatial transcriptomics. Science 353, 78-82 (2016)) (now available commercially as Visium); Visium Spatial Capture Technology, 10× Genomics, Pleasanton, CA; WO2020047007A2; WO2020123317A2; WO2020047005A1; WO2020176788A1; and WO2020190509A9), Slide-seq (Rodriques, S. G. et al. Slide-seq: A scalable technology for measuring genome-wide expression at high spatial resolution. Science 363, 1463-1467 (2019)), High Definition Spatial Transcriptomics (Vickovic, S. et al. High-definition spatial transcriptomics for in situ tissue profiling. Nat. Methods 16, 987-990 (2019)), SPLiT-seq (Rosenberg et al., Single-cell profiling of the developing mouse brain and spinal cord with split- pool barcoding. Science 360, 176-182 (2018)), or Chromium™ GEM-X (10x Genomics). Any of the methods described herein may include using the machine learning model to integrate imaging data, single cell sequencing data (e.g., single cell expression profile data), and in situ sequencing data (e.g., in situ expression profiling data). Example 1. Massively Scalable Drug Discovery Through Chemical Imaging and Machine Learning Referring to FIG.1, presented here is a technology by developing genetic screens, which may be used with studying, e.g., transcription factor, CRISPR, using various chemical imaging methods in both arrayed and pooled formats. Moreover, an integrated data analysis system has been constructed, capable of predicting perturbation effects, dosages, combinations, potential targets, and molecular mechanisms of perturbations. This allows an integrated experimental and computational solution for massively scalable mapping of phenotypes resulting from genetic perturbations on a large scale, incorporating chemical imaging, genetic screening, in situ sequencing, and machine learning. This represents a paradigm shift in drug discovery, distinct from all other existing technologies in the pharmaceutical industry. In particular, presented are: 1. A new experimental platform technology for mapping cell phenotypes after perturbation, which is leveraged for large-scale drug discovery. 2. A new computational platform technology for analyzing and predicting perturbation effects, dosages, combinations, mechanisms of action, and molecular mechanisms, enabling comprehensive understanding of molecular mechanisms. 3. An integrated drug discovery platform technology that facilitates large-scale mapping and prediction of cellular phenotypes post perturbation at low cost and high throughput. This comprehensive solution represents a novel modality in drug screening and discovery, distinct from existing approaches, and is easily adaptable to various perturbation assays such as small molecules, CRISPRs, TFs, mRNAs, peptides, and ADCs. Single-cell and spatial multi-omics are destructive: Single cell RNA-Seq (scRNA-seq) and other profiling assays (e.g., single-cell ATAC-seq, and single-cell proteomics) have opened new windows into understanding the properties, regulation, dynamics, and function of cells at unprecedented resolution and scale. However, these assays are inherently destructive, precluding us from tracking the temporal dynamics of live cells, in tissues, whole organisms, or humans. Non-destructive high-dimensional chemical imaging at single-cell resolution: There is a wide variety of biological imaging modalities, using contrast mechanisms such as nuclear magnetic resonance, Attorney Docket No.: 51812-006WO2 PATENT positron emission, fluorescence, ultrasound, etc. Among them, chemical imaging offers a unique capability to create a spatial map of chemical components (such as protein, lipids, DNA, and carbohydrate) in the sample. Briefly, every chemical compound displays a characteristic vibrational spectrum that is originated from interactions between light and the chemical structures, either via Raman scattering (excited by visible laser) or via infrared (IR) absorption (excited by mid-infrared light source). With the total/integrated vibrational spectrum of a collection of chemicals in a pixel, it is possible to “unmix” the total spectrum to individual spectral components and retrieve the concentration of individual chemicals. Spatial maps of chemicals are subsequently generated when collecting many pixels either via point scanning or wide-field microscopy measurement. Chemical imaging exhibits a notable uniqueness as it measures a high-dimensional property (a proxy to the biochemical phenotype) of the sample at every pixel. Either Raman spectrum or infrared spectrum, typically spanning from 500 to 3500 cm^-1 with ~2 cm^-1 spectral resolution of the standard instrument, contains thousands of independent spectral points. More specifically, the spectral region with rich biochemical information is known as the fingerprint region (900 to 1800 cm^-1) in the infrared spectrum, hence around 500 dimensions/channels per pixel (vs. fluorescent microscopy can usually generate less than 5 dimensions per pixel). When taking spatial information from each pixel within a cell into consideration, the information becomes enormously high. Such spatially resolved high-dimensional measurement will be key to our proposed technology. Results: Described below in Example 2. Methods: Experimental System: To demonstrate the generalizability of the technologies described herein, human embryonic stem cell (hESC) differentiation were used as an example. Understanding the intricate regulatory networks governing human embryonic stem cell (hESC) differentiation is crucial for advancing regenerative medicine and developmental biology. Transcription factors (TFs) play pivotal roles in orchestrating gene expression dynamics during differentiation. Integrating advanced imaging techniques like infrared imaging, Raman imaging, and stimulated Raman scattering (SRS) imaging with single-cell and in situ sequencing and machine learning holds promise for mapping TF responses with high resolution. Perturbation Assays: Genetic perturbation assays are experimental techniques used to manipulate genes or genetic elements in cells or organisms to study their function, interactions, and effects on various biological processes. This includes gene overexpression by such as transcription factor (TF) and other genes, gene knockdown by such as CRISPR, RNAi, gene editing such as CRISPR, base editing, prime editing. A TF screen in hESC was used as an example. A list of 108 TFs was identified in arrayed format and is used to generate a library of ~3000 TFs in pooled format.108 TFs are applied individually and combinatorially in arrayed format on hESC and map the genetic perturbation effects using vibrational imaging analysis and single cell sequencing profiling. For the whole ~3000 TF library, a pooled screen on hESC is performed Attorney Docket No.: 51812-006WO2 PATENT individually and combinatorially and the genetic perturbation effects are mapped using vibrational imaging analysis and single cell sequencing profiling, in particular, in situ sequencing to read out the genetic sequence of each TF in situ and to correlate the TF information with perturbation effects. Vibrational Imaging Analysis: Infrared Imaging: Visualize changes in chemical composition changes during hESC differentiation using infrared imaging. Identify chemical composition changes within differentiating cell populations. Infrared imaging provides extremely fast speed in imaging. Raman Imaging: Complementary to infrared imaging, Raman imaging provides high resolution in live cells. Visualize changes in chemical composition changes during hESC differentiation using Raman imaging. Identify chemical composition changes within differentiating cell populations at subcellular resolution in live cells. Raman imaging provides high spatial resolution in live cells over time. SRS Imaging: Infrared and Raman imaging capture the full spectrum, while SRS can map specific and biological relevant peaks. Capture real-time dynamics of TF perturbation responses using SRS imaging, enabling the visualization of TF perturbation activity with high spatiotemporal resolution and speed. Single-Cell Sequencing Profiling: scRNA-seq Data Analysis: Analyze single-cell multi-omics sequencing data to characterize gene regulation profiles associated with different stages of hESC differentiation upon TF perturbations. Identify TFs dynamically regulated during lineage specification. In Situ Sequencing: in situ sequencing to read out the genetic sequence of each TF in situ and correlate the TF information with perturbation effects. Integrate in situ sequencing data of genetic with vibrational imaging to correlate TF perturbation effects with gene regulation and chemical composition. Machine Learning Integration: The imaging and single cell sequencing data are analyzed to build machine learning models to predict perturbation effects, dosages, combinations, potential targets, and molecular mechanisms of perturbations. Supervised Learning Predictions: Train supervised learning models to predict TF responses based on combined vibrational imaging and single-cell sequencing data. Validate model predictions using experimental assays and imaging techniques. Unsupervised Learning Insights: Apply unsupervised learning algorithms to uncover latent structures in multidimensional datasets, revealing novel TF combinations and regulatory pathways involved in hESC differentiation. Integrating vibrational imaging, single-cell sequencing, and machine learning provides a powerful framework for mapping genetic perturbation responses. This multidisciplinary approach offers unprecedented resolution and insights into the regulatory networks governing cellular states. These platform technologies are highly modular and generalizable to different biological systems, genetic perturbation systems, phenotypic readout systems, and machine learning systems. The technologies Attorney Docket No.: 51812-006WO2 PATENT serve as a foundation for large-scale perturbation screens to identify novel perturbation strategies to convert cell fates, identify novel drug targets and understand disease mechanisms. Example 2. Massively Scalable Drug Discovery Through Chemical Imaging and Machine Learning An omics-imaging integrated screening method was used to decode the function of first-trimester placenta-related Transcription Factors (TFs) and identify a TFs combination that orchestrates the early development of the human placenta as a model system (FIG.2). The goals of this method were: 1. To establish an image-omics TF screening platform for studying their functions in early placenta development. 2. To develop a machine learning model that predicts omics changes based on TF perturbation and IR images. 3. To predict and validate the key/best TF cocktails for early placenta development. The MORF (multiplexed overexpression of regulatory factor) library is a comprehensive, barcoded ORF library focusing on regulatory transcription factors (TFs) involved in cell fate determination and differentiation (addgene.org). This library was used to select the TFs analyzed herein. Trophoblast differentiation from hESCs using TF perturbation was used as a model system. Based on in-house single-cell RNA-seq data of the human placenta, 108 TFs (Table 1) were selected. These 108 TFs are highly expressed in the human first-trimester placenta, indicating their potential roles in trophoblast lineage differentiation. Table 1. A list of 108 TFs that are potential regulators for trophoblast differentiation. Attorney Docket No.: 51812-006WO2 PATENT Following this, lentiviruses were packaged with the 108 TFs and were used to infect human embryonic stem cells (hESCs) in an arrayed format. To meet the FTIR imaging requirements, the cells were seeded onto CaF₂ slides within 24-well chambers. Experiments have been completed for 32 TFs, ARNTL, ZNF670, TP63, GRHL3, DACH1, DACH1, MEF2C, GATA3, ARNTL, TFAP2C, MECOM, TP63, GATA3, TEAD4, TCF4, SOX5, BRIP1, ARNTL, CTNNB1, ZEB1, SOX5, AFF2, GATA3, AFF2, ARNTL, TP63, GATA3, MECOM, ARNTL, CTNNB1, RFX6, and UHRF1, and the results are described in FIG.3. To achieve both single-cell multi-omics analysis and FTIR imaging simultaneously, two parallel groups of cell samples were designed. After inducing early differentiation in hESCs, single-cell multi- omics profiling and FTIR imaging analysis were performed on each sample. Hierarchical clustering of TIR mean spectra was performed (FIG.3. Proof of FTIR spectrum changes under different TF treatments). Other Embodiments While the invention has been described in connection with specific embodiments thereof, it will be understood that it is capable of further modifications and this application is intended to cover any variations, uses, or adaptations of the invention following, in general, the principles of the invention and including such departures from the invention that come within known or customary practice within the art to which the invention pertains and may be applied to the essential features hereinbefore set forth, and follows in the scope of the claims. Other embodiments are within the claims.

Claims

Attorney Docket No.: 51812-006WO2 PATENT CLAIMS 1. A method of analyzing a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data. 2. A method of identifying a biological state in a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates the biological state. 3. A method of determining the efficacy of an agent in effecting a change in a biological sample comprising: (a) obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) obtaining sequencing data of the biological sample after administration of the agent wherein the sequencing data are obtained using single-cell sequencing and/or in situ sequencing; and (c) applying, by a computer, a machine learning model to the one or more images and the sequencing data to correlate the one or more images with the sequencing data, wherein a correlation between the one or more images and the sequencing data indicates that the agent is effective in effecting a change in the biological sample. 4. A method of analyzing a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to the one or more images and the sequencing data to analyze the one or more images. 5. A method of identifying a biological state in a biological sample comprising: (a) obtaining one or more images of the biological sample using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to determine the biological state. Attorney Docket No.: 51812-006WO2 PATENT 6. A method of determining the efficacy of an agent in effecting a change in a biological sample comprising: (a) obtaining one or more images of the biological sample after administration of the agent, wherein the one or more images are obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) applying, by a computer, a machine learning model to determine that the agent is effective in effecting a change in the biological sample. 7. The method of any one of claims 1-6, wherein the machine learning model was developed using a combination of: (a) an imaging technique comprising infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) a sequencing technique comprising single-cell sequencing and/or in situ sequencing. 8. The method of any one of claims 1-7, wherein the machine learning model is a supervised learning model, an unsupervised learning model, a self-supervised learning model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model. 9. The method of any one of claims 1-8, wherein each of the one or more images comprises a collection of pixels, wherein each pixel comprises an infrared spectrum, a Raman spectrum, or a combination thereof. 10. The method of claim 9, wherein the method further comprises determining a chemical composition of the biological sample at each pixel. 11. The method of any one of claims 1-10, wherein the one or more images are 3D images. 12. The method of claim 11, wherein the 3D images comprise optical sections, and wherein the optical sections comprise a thickness of 5 µm to 500 µm. 13. The method of claim 1-12, wherein the infrared microscopy produces a spectrum from 400 to 4000 cm^-1and/or the Raman microscopy produces a spectrum from 500 to 3500 cm^-1. 14. The method of claim 13, wherein the spectrum is from 900 to 1800 cm^-1. 15. The method of any one of claims 3 and 6-14, wherein the agent is a small molecule, a nucleic acid, a protein, a gene editing agent, or a cell. Attorney Docket No.: 51812-006WO2 PATENT 16. The method of claim 15, wherein the nucleic acid is an RNAi, an shRNA, an mRNA, or an antisense oligonucleotide. 17. The method of claim 15, wherein the protein is a transcription factor, an antibody, a peptide, a cytokine, a hormone, an enzyme, an antibody drug conjugate, or a fusion protein. 18. The method of claim 15, wherein the gene editing agent is clustered regularly interspaced short palindromic repeats (CRISPR), a zinc finger nuclease, a base editor, a prime editor, or transcription activator-like effectors (TALEN). 19. The method of claim 15, wherein the cell is a chimeric antigen receptor T (CAR-T) cell. 20. The method of any one of claims 1-19, wherein the biological sample comprises a cell, a tissue, and/or a cellular component. 21. The method of any one of claims 1-20, wherein the biological sample was obtained from a human. 22. The method of any one of claims 1-21, wherein the biological sample is obtained from a subject with a disorder. 23. The method of claim 21, wherein the biological sample comprises human embryonic stem cells. 24. The method of any one of claims 3 and 6-23, wherein the effected change comprises treating a disorder, altering a gene or protein expression profile, inducing cell differentiation, inducing cell death, or a combination thereof. 25. A computer whose input data is: (a) images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; or (b) images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing, wherein the computer is programmed with a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing. Attorney Docket No.: 51812-006WO2 PATENT 26. The computer of claim 25, wherein the machine learning model is a supervised learning model, an unsupervised learning model, a self-supervised learning model, a semi-supervised learning model, a contrastive learning model, a generative learning model, or a reinforcement learning model. 27. A system comprising: (a) an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; (b) a single-cell profiling component configured for single-cell sequencing and/or in situ sequencing; and (c) a computer configured to receive images from the imaging component and sequencing data from the single-cell profiling component, wherein the imaging component and the single-cell profiling component are each operatively coupled to the computer, wherein the computer is programed with a machine learning model whose input data is the images and the sequencing data, and wherein the machine learning model correlates the images with the sequencing data. 28. A system comprising: (a) an imaging component configured for infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and (b) a computer configured to receive images from the imaging component, wherein the imaging component is operatively coupled to the computer, and wherein the computer comprises a machine learning model developed using a combination of images obtained using infrared microscopy, Raman microscopy, stimulated Raman scattering (SRS) microscopy, or a combination thereof; and sequencing data obtained using single-cell sequencing and/or in situ sequencing.