Methods for analyzing cells

Abstract

The single cell transition feature of the difference cell component expression between the cell state representing the first cell state and the changed cell state is accessed.The changed state occurs by the cell transition from the first cell state to the changed cell state.The transition feature comprises the identification of multiple cell components and, for each such cell component, the corresponding first significant score of the association between the change of the expression of the corresponding cell component and the change of the cell state between the first cell state and the changed cell state is quantified.The transition feature is compared with the disturbance feature of the difference cell component expression between the disturbed cell representing the unperturbed cell and the disturbed cell exposed to the disturbance.The disturbance feature comprises, for each corresponding cell component, the corresponding second significant score of the association between the change of the expression of the cell component between (i) the unperturbed cell and the disturbed cell and (ii) the change of the cell state between the unperturbed cell and the disturbed cell is quantified.

Description

[0001] CROSS-REFERENCE TO RELATED APPLICATIONS

[0002] This application claims the benefit of priority to U.S. Patent Application No. 62 / 698,701 filed on July 16, 2018, U.S. Patent Application No. 62 / 805,884 filed on February 14, 2019, and U.S. Patent Application No. 62 / 805,888 filed on February 14, 2019, the entire contents of which are incorporated herein by reference. Technical Field

[0003] The present invention generally relates to systems and methods for analyzing cells. More particularly, the present invention relates to predicting whether a perturbation will affect a cell transition. Background Art

[0004] The study of cellular mechanisms is important for understanding disease.

[0005] Tissues are complex ecosystems of individual cells, in which dysregulation of cellular states underlies disease. Existing drug discovery efforts attempt to characterize the molecular mechanisms that cause cells to transition from a healthy to a diseased state and to identify pharmacological approaches to reverse or inhibit these transitions. Past efforts have also attempted to identify molecular signatures that characterize these transitions and to identify pharmacological approaches to reverse these signatures.

[0006] Molecular data from large collections of cells in tissues or cells rich in surface markers obscure the phenotypic and molecular diversity of individual cells in the population. The heterogeneity of cells in these large collections of cells causes the results of current work aimed at elucidating disease-driving mechanisms to be misleading or even completely incorrect. New methods, such as single-cell RNA sequencing, can characterize individual cells at the molecular level. These data provide the basis for understanding different cell states at a higher resolution and reveal the rich and significant diversity of states that cells have.

[0007] Significant challenges in interpreting single-cell data are the sparse nature of these data, ignoring the presence of molecules present in the cell, and noise, which introduces uncertainty into the accuracy of these molecular measurements. Therefore, new approaches are needed to gain insight into the pharmacological approaches that control the state of individual cells and, in turn, address disease.

[0008] Computational localization and relocalization of chemicals (including small molecules, extracellular ligands, mRNA, siRNA, and others) has great potential to accelerate drug discovery. Past approaches have mapped differential expression signatures derived from a population of cells perturbed by small molecules as cellular expression deltas between healthy and diseased states. This approach has potential, but its applicability in its current form is limited by the heterogeneity of the population of cells and the significant cell type differences between molecularly perturbed and diseased cells.

[0009] In view of the above background, there is a need in the art for systems and methods that can achieve enhanced cell analysis. In particular, there is a need to be able to predict whether a perturbation will affect cell transformation. Summary of the invention

[0010] The present disclosure addresses the above identified shortcomings. The present disclosure addresses these shortcomings at least in part by using single cell data and molecular perturbation data as key data foundations and using machine learning to improve the understanding of naturally diverse cell states, revealing key transition states when cells choose alternative states, promoting the understanding of the molecular mechanism basis of cell state changes and discovering pharmacological methods for controlling these state changes.

[0011] One aspect of the present disclosure provides a method for predicting whether a disturbance will affect a cell transition (e.g., whether to promote transition or inhibit transition). The method includes accessing a single cell transition feature in electronic form. The transition feature represents a measure of differential cell component expression between a first cell state and a changed cell state. The changed cell state occurs by a cell transition from a first cell state to a changed cell state. The single cell transition feature includes the identification of a plurality of cell components. For each corresponding cell component in the plurality of cell components, the corresponding first significant score quantifies the association between the change in expression of the corresponding cell component and the change in the cell state between the first cell state and the changed cell state. In fact, any number of single cell transition features can be obtained in this manner, each single cell transition feature representing a measure of differential cell component expression between a first cell state and different changed cell states. Therefore, the disclosed content of the present application can be used to analyze any number of different changed cell states in parallel.

[0012] The method also includes accessing disturbance features in electronic form. In some embodiments, the disturbance feature represents the measure of differential cell component expression between one or more unperturbed cells and one or more disturbed cells exposed to disturbance. In addition, the disturbance feature includes all or part of the identification of multiple cell components. For each corresponding cell component in all or part of multiple cell components, the corresponding second significant score quantifies the association between the expression change of the corresponding cell component between one or more unperturbed cells and one or more disturbed cells and the cell state change between one or more unperturbed cells and one or more disturbed cells. In fact, any number of disturbance features can be obtained in this way, and each disturbance feature represents the measure of differential cell component expression between one or more unperturbed cells and one or more disturbed cells exposed to different disturbances in multiple disturbances. In addition, the method includes comparing one or more single cell transition features and one or more disturbance features, thereby determining whether one or more disturbances will affect the transition of cells to the state of one or more changes. In some embodiments, two, three, four, ten or more (e.g., 15, 20, 25, 30, 40, 50, 60, 70, 80, 90, or 100 or more) altered states are analyzed in parallel in this manner. In some embodiments, two, three, four, ten or more (e.g., 15, 20, 25, 30, 40, 50, 60, 70, 80, 90, or 100 or more) perturbations are analyzed in parallel in this manner.

[0013] In some embodiments, accessing the single-cell transition signature comprises determining the single-cell transition signature based on a first plurality of first single-cell cell component expression datasets and a second plurality of second single-cell cell component expression datasets. Each corresponding first single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets is obtained from a corresponding single cell of the first plurality of cells in a first cell state. In addition, each corresponding second single-cell cell component expression dataset in the second plurality of second single-cell cell component expression datasets is obtained from a corresponding single cell of the second plurality of cells in an altered cell state.

[0014] In some embodiments, each corresponding data set of the first plurality of single cell cell component expression data sets includes a corresponding cell component vector in the first plurality of cell component vectors. In addition, each corresponding data set of the second plurality of single cell cell component expression data sets includes a corresponding cell component vector in the second plurality of cell component vectors. In addition, each corresponding cell component vector in the first plurality of cell component vectors and the second plurality of cell component vectors includes a plurality of elements. Each corresponding element in the corresponding cell component vector is associated with a corresponding cell component in the plurality of cell components and includes a corresponding value representing the amount of the corresponding cell component of the corresponding single cell, and the corresponding single cell is represented by a corresponding data set of the first plurality of single cell cell component expression data sets and the second plurality of single cell cell component expression data sets.

[0015] In some embodiments, the first plurality of single cell cell component expression data sets and / or the second plurality of single cell cell component expression data sets are performed with dimension reduction to generate a plurality of dimension reduction components. Therefore, for each corresponding cell component vector in the first plurality of cell component vectors and the second plurality of cell component vectors, the plurality of dimension reduction components are applied to the corresponding cell component vectors to form corresponding dimension reduction vectors, and the corresponding dimension reduction vectors include the dimension reduction component values ​​of each corresponding dimension reduction component in the plurality of dimension reduction components. This forms the corresponding first plurality of dimension reduction vectors and the second plurality of dimension reduction vectors. The method includes performing clustering to generate a set of clustering Cj. Each cluster includes a plurality of points corresponding to the subset of the first plurality of dimension reduction vectors and the second plurality of dimension reduction vectors. The first plurality of cells are identified from the first clustering of the set of clustering Cj, and the second plurality of cells are identified from the second clustering of the set of clustering Cj.

[0016] In some embodiments, manifold learning is performed using the corresponding first plurality of reduced dimensionality vectors and the second plurality of reduced dimensionality vectors to identify a relative cellular state of each cell in the first plurality of cells and the second plurality of cells relative to each other cell.

[0017] In some embodiments, the plurality of unperturbed cells are control cells that have not been exposed to the perturbation, or the unperturbed cells are an average of unrelated perturbed cells that have been exposed to the perturbation.

[0018] In some embodiments, the method further comprises pruning the single-cell transition signature and the perturbation signature to limit the plurality of cellular components to transcription factors.

[0019] In some embodiments, determining the single cell transition signature comprises determining the difference in the amount of cellular components in the plurality of cellular components between the first plurality of first single cell cellular component expression datasets and the second plurality of second single cell cellular component expression datasets using one of a difference of means test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model.

[0020] In some embodiments, the measure of differential cellular component expression quantifies the difference in cellular component amount between a third plurality of third single cell cellular component expression datasets and a fourth plurality of fourth single cell cellular component expression datasets using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model. Each corresponding third single cell cellular component expression dataset in the third plurality of third single cell cellular component expression datasets is obtained from a corresponding single cell of the plurality of unperturbed cells, and each corresponding fourth single cell cellular component expression dataset in the fourth plurality of fourth single cell cellular component expression datasets is obtained from a corresponding single cell of a fourth plurality of cells in the plurality of perturbed cells exposed to a perturbation.

[0021] In some embodiments, the single cell transition signature and the perturbation signature are filtered to reduce the number of cellular components included in the single cell transition signature and the perturbation signature. In some embodiments, filtering the single cell transition signature and the perturbation signature comprises reducing the number of cellular components included in the single cell transition signature and the perturbation signature according to a threshold p value or according to a threshold number of cellular components.

[0022] In some embodiments, for each respective cellular component in the plurality of cellular components, determining the corresponding second significance score for the respective cellular component comprises replacing the significance score for the respective cellular component with the corresponding match score for the respective cellular component. The match scores for the plurality of cellular components are combined to generate a match score for the perturbation. Based on the match scores for the respective perturbations, it is determined whether the perturbation is associated with a transition of the cell between the first cellular state and the altered cellular state.

[0023] In some embodiments, the corresponding matching scores include discrete or continuous scores.

[0024] In some embodiments, replacing the significant score includes replacing the significant score with a first score if the amount of the cellular component from the single cell transition feature of the corresponding cellular component and the amount of the cellular component from the perturbation feature of the corresponding cellular component are both upregulated. If the amount of the cellular component from the single cell transition feature of the corresponding cellular component is upregulated and the amount of the cellular component from the perturbation feature of the corresponding cellular component is downregulated, replacing the significant score with a second score. In addition, if the amount of the cellular component from the perturbation feature of the corresponding cellular component is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0025] In some embodiments, replacing a significant score includes replacing a significant score with a first score if the amount of a cell component from a single cell transition feature of the corresponding cell component and the amount of a cell component from a perturbation feature of the cell component are both down-regulated compared to a counterpart (e.g., a first cell state and an unperturbed state, respectively). If the amount of a cell component from a single cell transition feature of the corresponding cell component is down-regulated and the amount of a cell component from a perturbation feature of the cell component is up-regulated compared to a counterpart (e.g., a first cell state and an unperturbed state, respectively), replacing the significant score with a second score. In addition, if the amount of a cell component from a perturbation feature of the cell component is not significantly up-regulated or down-regulated compared to a counterpart (e.g., a first cell state and an unperturbed state, respectively), replacing the significant score with a third score.

[0026] In some embodiments, the plurality of cellular components includes a plurality of genes.

[0027] In some embodiments, each single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets and the second plurality of second single-cell cell component expression datasets is generated using a method comprising single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, or a combination or summary thereof.

[0028] In some embodiments, the method further comprises identifying the perturbation as one that promotes the altered cellular state based on the comparison.

[0029] In some embodiments, the cell transition signature and the perturbation signature are generated using different types of cellular components. In some embodiments, the cell transition signature and the perturbation signature are generated using the same type of cellular components.

[0030] In some embodiments, accessing in electronic form is performed for each corresponding feature in the plurality of perturbations, thereby obtaining a plurality of perturbation signatures. In addition, the comparing compares the single-cell transition signature and the perturbation signature to each corresponding feature in the plurality of perturbation signatures, thereby determining a subset of the plurality of perturbations associated with the transition of the cell between the first cell state and the altered cell state.

[0031] Another aspect of the present disclosure provides a method comprising accessing a plurality of single cell cell component expression datasets. Each single cell cell component expression dataset is obtained from a cell in a plurality of cells that have been transformed from the same "progenitor" cell type. Each dataset comprises a vector r of cell components. i . A vector of cell components r i Each entry in is associated with one of the plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell. The method also includes performing dimensionality reduction of the data set to generate a matrix M (e.g., a plurality of dimensionless components, such as Figure 1 The matrix M includes rows in the first dimension and columns in the second dimension. Each row corresponds to one of the plurality of cells. The values ​​of the matrix M include values ​​generated from the amounts of the cell components at points located in the first dimensional space and the second dimensional space. The method further includes performing clustering to generate clusters C j Each cluster includes a plurality of points corresponding to a subset of rows in the matrix M, and their corresponding cells. The method further includes using clusters C j The set of cells that differentially express cellular components E k of collection.

[0032] In certain embodiments, the method further comprises performing manifold learning under the relative similarity approximation of the points with the matrix M to create a matrix N. The matrix N comprises a plurality of rows (rows identical to the rows of the matrix M) and two columns. Each row corresponds to one of the plurality of cells, and each of the two columns corresponds to one of the two dimensions in the two-dimensional space. Based on the data set, the values ​​of the matrix N indicate the relative cell type of each cell relative to each other cell.

[0033] There are multiple embodiments of the multiple cells from which data sets are obtained. In certain embodiments, when obtaining a single cell cell component expression data set, the multiple cells are heterogeneous cell groups with various cell types. In additional embodiments, the multiple cells are homogeneous cell groups with "ancestor" cell types, and single cell cell component expression data sets are obtained at each of the multiple time points when the cell is transformed from the "ancestor" cell type, so that different data sets in the multiple data sets are collected for each unique cell and time point combination. In such embodiments, the multiple time points may include at least three time points. In further embodiments, the multiple time points may include the "ancestor" time points at which a considerable portion of the multiple cells have not yet transformed from the "ancestor" cell type. In some additional embodiments, the multiple time points may include the transition time points at which a considerable portion of the multiple cells have transformed from the "ancestor" cell type. In some further embodiments, the multiple time points may include at least one intermediate time point at which a considerable portion of the cells have at least partially transformed from the "ancestor" cell type.

[0034] The various cell components may also vary. For example, in some embodiments of the methods disclosed herein, the various cell components are selected from the group consisting of: nucleic acids, proteins, lipids, carbohydrates, nucleotides, and any combination thereof. In such embodiments, nucleic acids may be selected from the group consisting of DNA and RNA. In further embodiments, RNA may be selected from the group consisting of coding RNA and non-coding RNA. In certain embodiments, the multiple single cell cell component expression data sets are generated using a method selected from the group consisting of: single cell ribonucleic acid (RNA) sequencing (scRNA-seq), scTag-seq, single cell assay of transposase accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination or summary thereof.

[0035] Dimensionality reduction can be performed on the data set to generate reduced-dimensionality components (e.g., in the form of the above-mentioned matrix M) in a variety of ways. In certain embodiments, performing dimensionality reduction includes performing principal component analysis (PCA) on the single-cell cellular component expression data set to generate reduced-dimensionality components. In further embodiments, diffusion mapping and / or neural network autoencoders can be used to perform dimensionality reduction on the data set to generate reduced-dimensionality components.

[0036] Similarly, manifold learning can be performed using the reduced dimensionality components (e.g., in the form of a matrix M) to create another data form, such as a matrix N, in a variety of ways. In some embodiments of the present disclosure, performing manifold learning may include estimating the geometry of the data in the matrix M to create the matrix N. In such embodiments, performing manifold learning may include performing local linear embedding, local linear isometric mapping (ISOMAP), t-distributed stochastic neighbor embedding (t-SNE), Potential of Heat-Diffusion for Affinity Based Trajectory Embedding (PHATE), or unified manifold approximation and projection (UMAP). In further embodiments, performing manifold learning may include creating a force-directed layout based on the data in the matrix M to generate the matrix N. In one embodiment, the force-directed layout may be created using the Force Atlas 2 algorithm.

[0037] Clustering can also be performed in a variety of different ways. In some embodiments, performing clustering assumes no prior knowledge about the organization of the multiple points in each cluster. In additional embodiments disclosed herein, performing clustering includes performing HDBSCAN and / or Louvain community detection to generate clusters C j In a further embodiment, performing clustering includes assigning each point to a cluster C based on the time point at which the single cell cellular component expression dataset associated with each point was collected. j In some embodiments, performing clustering includes analyzing the plurality of points using a diffusion path algorithm that assigns the points to the clusters based on a measure of the degree to which the points are extremities of the clusters.

[0038] To determine the differentially expressed cellular components E k For example, in one embodiment, for each cell component, for at least one cluster C j , the amount of the cellular component of the plurality of points in the at least one cluster can be compared with the amount of the cellular component of the plurality of points in at least one other cluster. Then, in response to the amount of the cellular component of the plurality of points in the at least one cluster being greater than the threshold level of the amount of the cellular component of the plurality of points in the at least one other cluster, the cellular component can be added to the differentially expressed cellular component E k In some embodiments, the at least one cluster may include cluster C jIn a further embodiment, the at least one other cluster may include cluster C j off-lineage clusters containing spots with undesirable cell types.

[0039] In a further embodiment, to determine the differentially expressed cellular components E k For each cellular component, for at least one cluster, a distance metric can be calculated between the amount of the cellular component at multiple points in at least one cluster and the amount of the cellular component at multiple points in at least one other cluster. Then, in response to the distance metric being statistically significant, the cellular component can be added to the differentially expressed cellular components E k of concentration.

[0040] In certain embodiments, the methods described herein may further comprise screening the differentially expressed cellular components E against a transcription factor database. k In such an embodiment of identifying a set of differentially expressed transcription factors, the method may further comprise the following steps: k The method further comprises performing empirical mode decomposition on the set of differentially expressed transcription factors to generate a pseudo-time representation of the data set; and identifying the set of differentially expressed transcription factors based on the pseudo-time representation.

[0041] In another aspect, the present disclosure provides a method comprising accessing a plurality of single cell cell component expression data sets. Each data set is obtained from a cell in a plurality of cells that have been transformed from the same "progenitor" cell type. Each data set comprises a vector r of cell components. i . A vector of cell components r i Each entry in is associated with one of the multiple cell components, and the value of each entry represents the amount of the cell component of the cell. The method also includes generating a kNN graph using a kNN algorithm and a single cell cell component expression data set, performing clustering to generate a cluster C j and using the clustering C j The set of cells that differentially express E k Each cluster includes a plurality of points, each point corresponding to a single cell cell component expression data set of a cell in a plurality of cells. In some embodiments, the differentially expressed cell component E is determined. k The set includes the determined cluster C j The distance measure between multiple points in .

[0042] In another aspect, the present disclosure provides a method comprising accessing a single cell transition signature, the single cell transition signature representing a measure of differential cell component expression between a first cell state and an altered cell state. The method further comprises accessing a perturbation signature, the perturbation signature representing a measure of differential cell component expression between an unperturbed cell not exposed to the perturbation and a perturbed cell exposed to the perturbation. The method further comprises determining whether the perturbation is associated with a transition of a cell between a first cell state and an altered cell state based on a comparison of the single cell transition signature and the perturbation signature.

[0043] In some embodiments, accessing the single-cell transition signature comprises determining the single-cell transition signature based on a first plurality of single-cell cell component expression datasets and based on a second plurality of single-cell cell component expression datasets, each first dataset being obtained from a cell in a first plurality of cells in a first cell state, and each second dataset being obtained from a cell in a second plurality of cells in an altered cell state. For each cell, each of the first plurality of single-cell cell component expression datasets and the second plurality of single-cell cell component expression datasets may include a vector r of cell components. i , each entry in the vector is associated with one of the plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell. In some embodiments, determining the single-cell transition signature based on the first plurality of single-cell cellular component expression datasets and the second plurality of single-cell cellular component expression datasets comprises using one of a mean difference test, a Wilcoxon rank sum test (Mann Whitney U test), a t-test, a logistic regression, and a generalized linear model to determine the difference in cellular component amounts between the first plurality of single-cell cellular component expression datasets and the second plurality of single-cell cellular component expression datasets.

[0044] In an embodiment in which such a single-cell transition signature includes determining a single-cell transition signature based on a first plurality of single-cell cell component expression data sets and a second plurality of single-cell cell component expression data sets, the method may further include obtaining the first plurality of single-cell cell component expression data sets and the second plurality of single-cell cell component expression data sets. The obtaining step further includes performing dimensionality reduction on the first plurality of single-cell cell component expression data sets and the second plurality of single-cell cell component expression data sets to generate a matrix M. The matrix M includes rows in a first dimension and columns in a second dimension. Each row of the matrix M corresponds to a cell in the plurality of cells. The values ​​of the matrix M include values ​​generated from the amounts of cell components at points located in the first dimensional space and the second dimensional space. Obtaining the first single-cell cell component expression data set and the second single-cell cell component expression data set also includes performing clustering to generate clusters C jEach cluster includes a plurality of points corresponding to a subset of rows in the matrix M, and their corresponding cells. Obtaining the first single-cell cell component expression data set and the second single-cell cell component expression data set even includes obtaining the expression data from cluster C. j The first cluster of the set identifies the first plurality of cells from cluster C j A second cluster of the set identifies a second plurality of cells, a first plurality of single-cell cellular component expression datasets are obtained from the first plurality of cells, and a second plurality of single-cell cellular component expression datasets are obtained from the second plurality of cells.

[0045] In some embodiments, obtaining a first single-cell cell component expression data set and a second single-cell cell component expression data set further comprises performing manifold learning under a relative similarity approximation of points with the matrix M to create a matrix N. The matrix N comprises a plurality of rows and two columns. Each row corresponds to a cell in the first plurality of cells and the second plurality of cells, and each column corresponds to one or two dimensions in a two-dimensional space. Based on the first plurality of single-cell cell component expression data sets and the second plurality of single-cell cell component expression data sets, the values ​​of the matrix N indicate the relative cell state of each cell relative to each other cell.

[0046] In certain embodiments, accessing disturbance features may include determining disturbance features based on multiple unperturbed single-cell cell component expression data sets of unperturbed cells that are not exposed to the disturbance and based on multiple disturbed single-cell cell component expression data sets of disturbed cells that are exposed to the disturbance. Unperturbed cells may be control cells that are not exposed to the disturbance of the disturbed cells. Alternatively, unperturbed cells may be the average value of unrelated disturbed cells that have been exposed to the disturbance. In some embodiments, determining disturbance features based on multiple unperturbed single-cell cell component expression data sets and multiple disturbed single-cell cell component expression data sets may include using one of a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model to determine the difference in the amount of cellular components between multiple unperturbed single-cell cell component expression data sets and multiple disturbed single-cell cell component expression data sets.

[0047] In some embodiments, the method further comprises filtering the single cell transition signature and the perturbation signature to include cellular components that are transcription factors. In additional embodiments, the method further comprises filtering the single cell transition signature and the perturbation signature to reduce the number of cellular components included in the single cell transition signature and the perturbation signature. Specifically, the single cell transition signature and the perturbation signature may be filtered to reduce the number of cellular components included in the single cell transition signature and the perturbation signature according to a threshold p value or according to a threshold number of cellular components.

[0048] In further embodiments of the methods disclosed herein, the perturbation feature may include multiple cellular components, each cellular component being associated with a significant score, the significant score quantifying the association between the change in the amount of the cellular component and the change in the cell state between the unperturbed cell and the perturbed cell. In such embodiments, determining whether the perturbation is associated with the transition of the cell between the first cell state and the changed cell state may include: replacing the significant score of the cellular component with the matching score of each cellular component; combining the matching scores of the multiple cellular components to generate the matching score of the perturbation; and determining whether the perturbation is associated with the transition of the cell between the first cell state and the changed cell state based on the matching score of the perturbation. The matching score may include a discrete score or a continuous score. Replacing the significant score may include: if the amount of the cellular component from the single cell transition feature of the cellular component and the amount of the cellular component from the perturbation feature are both upregulated, replacing the significant score with the first score; if the amount of the cellular component from the single cell transition feature of the cellular component is upregulated and the amount of the cellular component from the perturbation feature is downregulated, replacing the significant score with the second score; and if the amount of the cellular component from the perturbation feature of the cellular component is not significantly upregulated or downregulated, replacing the significant score with the third score. Alternatively, replacing the significance score may include: if the amount of the cellular component from the single-cell transition feature and the amount of the cellular component from the perturbation feature of the cellular component are both downregulated, replacing the significance score with a first score; if the amount of the cellular component from the single-cell transition feature of the cellular component is downregulated and the amount of the cellular component from the perturbation feature is upregulated, replacing the significance score with a second score; and if the amount of the cellular component from the perturbation feature of the cellular component is not significantly upregulated or downregulated, replacing the significance score with a third score.

[0049] In an alternative embodiment of the method disclosed herein, the matching score is not used to replace the significant score associated with the cellular component of the disturbance feature. Instead, in an alternative embodiment, the disturbance feature can include multiple cellular components, each cellular component is associated with a significant score, and the significant score quantifies the association between the change in the amount of the cellular component and the change in the cell state between the unperturbed cell and the perturbed cell. In such an embodiment, determining whether the disturbance is associated with the transition of the cell between the first cell state and the changed cell state can include: simply combining the significant scores of the multiple cellular components to generate a significant score for the disturbance; and determining whether the disturbance is associated with the transition of the cell between the first cell state and the changed cell state based on the significant score of the disturbance.

[0050] In some embodiments, the false cell component discovery rate of the perturbed match score is estimated to determine the confidence level in the perturbation. In such embodiments, the false cell component discovery rate is estimated by: calculating the empirical marginal expression frequency of each cell component of a plurality of cell components; summing the empirical marginal expression frequencies of the plurality of cell components over their combination to generate the probability of identifying many cell components by assuming independent distribution of expression by chance; and estimating the false cell component discovery rate of the perturbed match score based on the probability.

[0051] In some embodiments, determining whether a perturbation is associated with a transition of a cell between a first cell state and an altered cell state depends on a covariate of the perturbation. For example, in some embodiments, determining whether a perturbation is associated with a transition of a cell between a first cell state and an altered cell state may include: determining that a threshold amount of a covariate of the perturbation is associated with a transition of a cell between a first cell state and an altered cell state; and in response to the determination, determining that the perturbation is associated with a transition of a cell between a first cell state and an altered cell state. In some embodiments, the perturbation may include exposing the cell to a small molecule. The covariate of the perturbation may include a specific dose of the small molecule, a time for measuring differential cellular component expression between an unperturbed cell and a perturbed cell relative to the time the perturbed cell is exposed to the small molecule, and a cell line of the perturbed cell.

[0052] In certain embodiments, the cellular component may include a gene. The single cell cellular component expression dataset may be generated using a method selected from the group consisting of single cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single cell assay for transposase accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination or summary thereof.

[0053] In some embodiments of the methods disclosed herein, at least one of the single cell transition signature and the perturbation signature is obtained from a database. The perturbation signature can be obtained from a database of multiple perturbation signatures including multiple perturbations. In such embodiments, for each perturbation in the multiple perturbations in the database, the perturbation signature of the perturbation is accessed from the database, and based on a comparison of the single cell transition signature with the perturbation signature, it is determined whether the perturbation is associated with a transition of the cell between the first cell state and the altered cell state.

[0054] In further embodiments of the methods disclosed herein, the methods may further include: accessing a plurality of perturbation signatures from a plurality of perturbed cells; and screening perturbations that promote an altered cell state by determining, for each of the plurality of perturbation signatures, whether a perturbation associated with the perturbation signature is associated with a transition of the cell between a first cell state and an altered cell state based on a comparison of the single cell transition signature and the perturbation signature. In embodiments of screening perturbation signatures that promote perturbations that promote an altered cell state, accessing a plurality of perturbation signatures may include: exposing cells to a plurality of perturbations to generate a plurality of perturbed cells; and measuring amounts of cellular components from the plurality of perturbed cells.

[0055] The method may also include identifying a disturbance that promotes a changed cell state. Promoting a changed cell state may include, in a cell population including a first cell state, promoting a transition from a first cell state to a changed cell state. Alternatively, promoting a changed cell state may include, in a cell population including a first cell state, increasing the ratio of the number of cells in an alternative state to the number of cells in a first state or a state that is optionally different from the changed cell state. In a further alternative embodiment, promoting a changed cell state may include, in a cell population including a first cell state, increasing the absolute number of cells in a changed cell state. In a further alternative embodiment, promoting a changed cell state may include, in a cell population including a first cell state, reducing the absolute number of cells in a first cell state or a state that is optionally different from the changed cell state.

[0056] In certain embodiments, the cell transition signature and the perturbation signature can be generated using different types of cellular components. For example, the cell transition signature can be generated based on RNA expression (e.g., counts of RNA transcripts), and the perturbation signature can be generated based on protein expression (e.g., counts of amino acids). In alternative embodiments, the cell transition signature and the perturbation signature can be generated using the same types of cellular components. For example, both the cell transition signature and the perturbation signature can be generated based on RNA expression (e.g., counts of RNA transcripts).

[0057] In another aspect, the present disclosure provides a method comprising accessing a single cell transition signature, the single cell transition signature representing a measure of differential cell component expression between a first cell state and an altered cell state. The method further comprises accessing a plurality of perturbation signatures, each perturbation signature associated with a perturbation representing a measure of differential cell component expression between an unperturbed cell not exposed to the perturbation and a perturbed cell exposed to the perturbation. The method further comprises determining a subset of perturbations associated with a transition of a cell between a first cell state and an altered cell state based on a comparison of the single cell transition signature with the plurality of perturbation signatures.

[0058] In certain embodiments, each disturbance feature includes multiple cellular components, and each cellular component is associated with a significant score, and the significant score quantifies the association between the change in the amount of the cellular component and the change in the cell state between the unperturbed cell and the perturbed cell. In such an embodiment, determining a subset of disturbances associated with the transition of a cell between a first cell state and a changed cell state includes, for each disturbance feature, replacing the significant score of the cellular component with the matching score of each cellular component, and combining the matching scores of the multiple cellular components to generate a matching score for the disturbance. Then, the method also includes sorting the disturbances according to the matching scores of the disturbances, and selecting a subset of the disturbances based on the sorted list of the disturbances.

[0059] In another aspect, the present disclosure disclosed herein provides a computer program product comprising a non-transitory computer-readable storage medium on which instructions are encoded. When executed by a processor, the encoded instructions cause the processor to perform any embodiment of the method disclosed herein. In yet another aspect, the invention disclosed herein provides a system comprising a non-transitory computer-readable storage medium on which instructions are encoded. When executed by a processor, the encoded instructions cause the processor to perform any embodiment of the method disclosed herein.

[0060] In yet another aspect, the present disclosure provides a method for promoting neurons and / or "progenitor" cells. The method comprises exposing a starting fibroblast population to a disturbance having a disturbance characteristic that promotes the transformation of the starting fibroblast population to "progenitor" cells and / or neurons. In such an embodiment, the disturbance characteristic is an increase in the activity of one or more of Brn2, Ascl1, Myt1, Zfp941, Taf5B, St18, Zkscan16, Camta1, and Arnt2 and / or a decrease in the activity of one or more of Ascl1, Atf3, Rorc, Scx, Satb1, Elf3, and Fos.

[0061] In certain embodiments of the method for promoting neurons and / or "progenitor" cells, the neurons and / or "progenitor" cells are promoted by one or more of the following: increasing the absolute number of neurons and / or "progenitor" cells; reducing the absolute number of fibroblasts; promoting the transformation of fibroblasts to neurons and / or "progenitor" cells; promoting the life span of neurons or "progenitor" cells; reducing the life span of fibroblasts; or increasing the ratio of neurons and / or "progenitor" cells to fibroblasts. In further embodiments, the perturbation does not include Forskolin, PP1, PP2, and Trichostatin A.

[0062] In another aspect, the present disclosure provides a method for increasing the amount of neurons and / or "progenitor" cells. The method includes exposing a fibroblast group to a pharmaceutical composition having a perturbation characteristic that promotes the transformation of a fibroblast group into neurons. The pharmaceutical composition comprises forskolin, PP1, PP2, trichostatin A, BRD-K38615104, Geldanamycin, Manumycin A, Mitoxantrone, Curcumin, Alvocidib, Varinostat, KI20227 or a combination of the foregoing substances, such as a combination of 2, 3, 4, 5 or more of the foregoing substances. In some embodiments, the pharmaceutical composition does not comprise forskolin, PP1, PP2 and trichostatin A.

[0063] In another aspect, the present disclosure provides a pharmaceutical composition for promoting neurons and / or "progenitor" cells. The pharmaceutical composition comprises a perturbation selected from the group consisting of forskolin, PP1, PP2, trichostatin A, BRD-K38615104, geldanamycin, manumycin A, mitoxantrone, curcumin, avosid, varinostat, KI20227 or a combination of the foregoing substances and a pharmaceutically acceptable excipient. In some embodiments, the perturbation does not include forskolin, PP1, PP2 and trichostatin A.

[0064] In yet another aspect, the present disclosure provides a unit dosage form comprising one of the pharmaceutical compositions disclosed herein.

[0065] In another aspect, the present disclosure provides a method for identifying a candidate perturbation for promoting the transformation of a starting fibroblast group into neurons and / or "progenitor" cells. The method includes exposing the starting fibroblast group to a perturbation and identifying a perturbation feature of the perturbation. The perturbation feature of the perturbation includes one or more cell components and a significant score associated with each cell component. The significant score of each cell component quantifies the association between the change in cell component expression after the fibroblast group is exposed to the perturbation and the change in the cell state of the fibroblast group to neurons and / or "progenitor" cells. The perturbation feature includes an increase in the activity of one or more of Brn2, Ascl1, Myt1, Zfp941, Taf5B, St18, Zkscan16, Camta1 and Arnt2 and / or a decrease in the activity of one or more of Ascl1, Atf3, Rorc, Scx, Satb1, Elf3 and Fos. The method also includes identifying the perturbation as a candidate perturbation for promoting the transformation of a fibroblast group into neurons and / or "progenitor" cells based on the perturbation feature.

[0066] Yet another aspect of the present disclosure provides a non-transitory computer-readable storage medium, wherein the non-transitory computer-readable storage medium stores instructions which, when executed by a computer system, cause the computer system to perform any of the methods for analyzing cells described in the present disclosure. BRIEF DESCRIPTION OF THE DRAWINGS

[0067] Embodiments disclosed herein are illustrated by way of example and not limitation in the figures of the accompanying drawings. Throughout the figures, like reference numerals refer to corresponding parts.

[0068] Figure 1 A block diagram illustrating an exemplary system and computing device according to one embodiment of the present disclosure;

[0069] Figure 2 A flow chart is provided of methods and features of a system for analyzing cells according to various embodiments of the present disclosure, wherein elements in dashed boxes are optional;

[0070] Figure 3 is a flow chart of a first example of a differential cellular component expression assay for determining a set of differentially expressed cellular components according to one embodiment of the present disclosure;

[0071] Figure 4A depicting a timeline tracking the trajectory of an induced cell state transition over a period of time according to one embodiment of the present disclosure;

[0072] Figure 4B depicts a manifold generated by a force-directed layout algorithm for the exemplary matrix N in Supplementary Table 1 according to one embodiment of the present disclosure;

[0073] Figure 5A Describes an embodiment of the present disclosure Figure 5B Manifolds;

[0074] Figure 5B A depiction of an embodiment according to the present disclosure is Figure 4B The expression level of each BAM transcription factor in each cell on each measurement day for each point in the manifold;

[0075] Figure 6 depicting images of MEF cells in which Ascl1 transcription factor expression is forced, which have been stained with DAPI, Map2 antibody, and Tuj1 antibody, images of mouse neurons stained with DAPI, Map2 antibody, and Tuj1 antibody, and images of MEF cells in which Ascl1 transcription factor expression is not forced, which have been stained with DAPI, Map2 antibody, and Tuj1 antibody, according to one disclosed embodiment;

[0076] Fig. 7ADescribes an embodiment of the present disclosure Figure 4B manifold where the points in the manifold are grouped into clusters C identified by clustering j middle;

[0077] Figure 7B Delineating transcription factors known and unknown in the literature associated with the transition of MEFs to mouse neurons (and vice versa to mouse myocytes) according to one embodiment of the present disclosure;

[0078] Fig. 8A Describes an embodiment of the present disclosure Figure 4A Mapping of the transformation trajectories of the MEF cells in question;

[0079] Figure 8B Describes a method for identifying a perturbation that affects a cell's transition trajectory by altering gene expression in the cell such that Fig. 8A The transition trajectory is mapped to achieve the transition of cells from the first state to the second state;

[0080] Fig. 9 Depicting small molecule perturbations associated with the transition of MEFs to mouse neurons (and vice versa to mouse myocytes) according to one embodiment of the present disclosure;

[0081] Fig. 10A providing a bar graph showing the total number of neurons under each treatment condition according to one embodiment of the present disclosure, wherein the total number of neurons is manually counted based on positive Tuj1 / Map2 signals and neuronal morphology, and wherein for each experiment, the data for each treatment condition is normalized by the number of neurons in DMSO-treated wells; and

[0082] Fig. 10B A bar graph showing the percentage of neurons under each treatment condition is provided according to one embodiment of the present disclosure. DETAILED DESCRIPTION

[0083] Reference will now be made in detail to the embodiments, examples of which are illustrated in the accompanying drawings. In the following detailed description, many specific details are set forth to provide a thorough understanding of the present disclosure. However, it will be apparent to those skilled in the art that the present disclosure may be practiced without these specific details. In other cases, well-known methods, processes, components, circuits, and networks are not described in detail to avoid unnecessary confusion of the various aspects of the embodiments.

[0084] Multiple instances may be provided for components, operations or structures described herein as single instances. Finally, the boundaries between various components, operations and data storage are arbitrary to some extent, and specific operations are described in the context of specific illustrative configurations. Other forms of functions are contemplated, and they may fall within the scope of one or more embodiments. Typically, the structures and functions presented as separate components in the exemplary configuration may be implemented as combined structures or components. Similarly, the structures and functions presented as separate components may be implemented as separate components. These and other variations, modifications, additions and improvements all fall within the scope of the one or more embodiments.

[0085] It should also be understood that although the terms "first", "second", etc. may be used herein to describe various elements, these elements should not be limited by these terms. These terms are only used to distinguish one element from another element. For example, without departing from the scope of the present invention, a first data set may be referred to as a second data set, and similarly, a second data set may be referred to as a first data set. The first data set and the second data set are both data sets, but they are not the same data set.

[0086] The terms used herein are only used for the purpose of describing specific embodiments and are not intended to limit the claims. As used in the description of the embodiments and the appended claims, unless otherwise clearly indicated in the context, the singular forms "one", "a kind" and "the" are intended to include the plural forms as well. It should also be understood that the term "and / or" as used herein refers to and encompasses any and all possible combinations of one or more associated listed items. It should also be understood that when used in this specification, the terms "comprising" and / or "including" indicate the presence of stated features, integers, steps, operations, elements and / or parts, but do not exclude the presence or increase of one / kind or more / kind of other features, integers, steps, operations, elements, parts and / or their groups.

[0087] As used herein, the term "if" may be interpreted to mean "when the stated precondition is true," or "when the stated precondition is true," or "in response to determining," or "according to determining," or "in response to detecting" that the stated precondition is true, depending on the context. Similarly, the phrase "if it is determined that (the stated precondition is true)" or "if (the stated precondition is true)" or "when (the stated precondition is true)" may be interpreted to mean "upon determining that the stated precondition is true," or "in response to determining that the stated precondition is true," or "according to determining that the stated precondition is true," or "in detecting that the stated precondition is true," or "in response to detecting that the stated precondition is true," depending on the context.

[0088] In addition, when a reference number is given as an "i" representation, the reference number refers to a common component, set or embodiment. For example, a cell component referred to as "cell component i" refers to the i-th cell component among a plurality of cell components.

[0089] The foregoing description includes exemplary systems, methods, techniques, instruction sequences, and computing machine program products that embody illustrative embodiments. For purposes of explanation, many specific details are set forth to provide an understanding of the various embodiments of the subject matter of the present invention. However, it will be apparent to those skilled in the art that the embodiments of the subject matter of the present invention may be practiced without these specific details. Typically, well-known instruction instances, protocols, structures, and techniques are not shown in detail.

[0090] For the purpose of explanation, the foregoing description has been described with reference to specific embodiments. However, the illustrative discussion below is not intended to be exhaustive or to limit the embodiments to the precise form disclosed. Many modifications or variations are possible in light of the above teachings. The embodiments are selected and described to best explain the principles and their practical application, thereby enabling others skilled in the art to best utilize the embodiments and various embodiments with various modifications suitable for the intended specific use.

[0091] For the sake of clarity, not all routine features of the embodiments described herein are shown and described. It should be understood that in the development of any such actual embodiment, many implementation-specific decisions are made to achieve the specific goals of the designer, such as compliance with use case-related and business-related constraints, and these specific goals vary from implementation to implementation and from designer to designer. In addition, it should be understood that such design work may be complex and time-consuming, but it is still a routine engineering task for those skilled in the art who benefit from this disclosure.

[0092] Certain portions of this description describe embodiments of the present invention in terms of algorithms and symbolic representations of operations on information. These algorithmic descriptions and representations are commonly used by those skilled in the art of data processing to effectively convey the substance of their work to others skilled in the art. Although these operations are described functionally, computationally, or logically, it is understood that these operations are implemented by computer programs or equivalent circuits, microcode, etc.

[0093] The language used in this specification is selected primarily for readability and instructional purposes, and may not be selected to describe or limit the subject matter of the present invention. Therefore, it is intended that the scope of the present invention is not limited by this specific embodiment, but by any claims proposed for the application based thereon. Therefore, the disclosure of the embodiments of the present invention is intended to illustrate rather than limit the scope of the present invention.

[0094] Generally, the terms used in the claims and description are intended to be interpreted as having the ordinary meaning as understood by one of ordinary skill in the art. Certain terms are defined below to provide additional clarity. If the ordinary meaning conflicts with the provided definition, the provided definition will apply.

[0095] Any term not directly defined herein should be understood to have a meaning generally associated with their understanding in the field of the present invention. Certain terms are discussed herein to provide additional guidance to practitioners in terms of compositions, devices, methods, etc. describing various aspects of the present invention and how to make or use them. It should be understood that the same thing can be stated in more than one way. Therefore, alternative language and synonyms can be used for any one or more of the terms discussed herein. It is not important whether the terms are elaborated or discussed herein. Some synonyms or alternative methods, materials, etc. are provided. Unless clearly stated, the listing of one or more synonyms or equivalents does not exclude the use of other synonyms or equivalents. Examples (including examples of terms) are used only for illustrative purposes and do not limit the scope and meaning of various aspects of the present invention herein.

[0096] As used herein, the term "perturbation" with respect to a cell (e.g., perturbation of a cell or cell perturbation) refers to any treatment of a cell with one or more compounds. These compounds may be referred to as "perturbagens". In some embodiments, perturbagens may include, for example, small molecules, biologics, proteins, proteins combined with small molecules, ADCs, nucleic acids (such as siRNA or interfering RNA), cDNAs overexpressing wild-type and / or mutant shRNAs, cDNAs overexpressing wild-type and / or mutant guide RNAs (e.g., Cas9 systems or other gene editing systems), or any combination of any of the foregoing.

[0097] As used herein, the term "progenitor" with respect to a cell (eg, a progenitor cell) refers to any cell that is capable of transitioning from one cellular state to at least one other cellular state.

[0098] As used herein, the term "dataset" for cell component expression measurements for one or more cells may refer to a high-dimensional data set collected from a single cell (e.g., a single-cell cell component expression data set) in some contexts. In other contexts, the term "dataset" may refer to a plurality of high-dimensional data sets collected from a single cell (e.g., a plurality of single-cell cell component expression data sets), each of the plurality of data sets being collected from a cell in a plurality of cells.

[0099] As used herein, the term "affect" refers to a change in cellular transformation.

[0100] I. Exemplary System Implementation

[0101] Now that an overview of some aspects of the disclosure and some definitions used in the disclosure have been provided, Figure 1 Details of an exemplary system are described.

[0102] Figure 1 A block diagram illustrating a system 100 according to some embodiments of the present disclosure is provided. The system 100 provides a prediction of whether a perturbation will affect cell transformation. Figure 1 In the example, system 100 is illustrated as a computing device. Of course, other topologies of computer system 100 are possible. For example, in some embodiments, system 100 may actually constitute several computer systems linked together in a network, or may be a virtual machine or container in a cloud computing environment. Figure 1 The exemplary topology shown in FIG. 1 is intended merely to describe features of one embodiment of the present disclosure in a manner that will be readily understood by those skilled in the art.

[0103] See also Figure 1 In some embodiments, the computer system 100 (e.g., computing device) includes a network interface 104. In some embodiments, the network interface 104 interconnects the system 100 computing devices within the system with each other and optionally external systems and devices through one or more communication networks (e.g., through a network communication module 118). In some embodiments, the network interface 104 optionally provides communication through the network communication module 118 via the Internet, one or more local area networks (LANs), one or more wide area networks (WANs), other types of networks, or a combination of such networks.

[0104] Examples of networks include the World Wide Web (WWW), an intranet and / or wireless networks, such as cellular telephone networks, wireless local area networks (LANs) and / or metropolitan area networks (MANs) and other means of communicating via wireless communications. Wireless communication optionally uses any of a number of communication standards, protocols and technologies, including Global System for Mobile Communications (GSM), Enhanced Data GSM Environment (EDGE), High Speed ​​Downlink Packet Access (HSDPA), High Speed ​​Uplink Packet Access (HSUPA), Evolution Data-Only (EV-DO), HSPA, HSPA+, Dual-Cell HSPA (DC-HSPDA), Long Term Evolution (LTE), Near Field Communication (NFC), Wideband Code Division Multiple Access (W-CDMA), Code Division Multiple Access (CDMA), Time Division Multiple Access (TDMA), Bluetooth, Wireless Fidelity (Wi-Fi) (e.g., IEEE 802.11a, IEEE 802.11ac, IEEE 802.11ax, IEEE 802.11b, IEEE 802.11g and / or IEEE 802.11n), Voice over Internet Protocol (VoIP), Wi-MAX, email protocols (e.g., Internet message access protocol (IMAP) and / or post office protocol (POST)). protocol; POP)), instant messaging (e.g., extensible messaging and presence protocol; XMPP), Session Initiation Protocol for Instant Messaging and Presence Leveraging Extensions; SIMPLE, Instant Messaging and Presence Service; IMPS, and / or Short Message Service; SMS)), or any other suitable communications protocol, including communications protocols that have not been developed as of the filing date of this document.

[0105] In some embodiments, the system 100 includes one or more processing units (CPUs) 102 (e.g., processors, processing cores, etc.), one or more network interfaces 104, a user interface 107 for use by a user including (optionally) a display 108 and an input system 110 (e.g., input / output interface, keyboard, mouse, etc.), memory (e.g., non-persistent memory 111, persistent memory 112), and one or more communication buses 114 for interconnecting the aforementioned components. The one or more communication buses 114 optionally include circuits (sometimes referred to as chipsets) that interconnect and control communications between system components. The non-persistent memory 111 typically includes high-speed random access memory, such as DRAM, SRAM, DDR RAM, ROM, EEPROM, flash memory, and the persistent memory 112 typically includes a CD-ROM, a digital versatile disk (DVD) or other optical storage device, a magnetic cassette, a magnetic tape, a magnetic disk storage device or other magnetic storage device, a magnetic disk storage device, an optical disk storage device, a flash memory device, or other non-volatile solid-state storage device. Persistent storage 112 optionally includes one or more storage devices located remotely from one or more CPUs 102. Persistent storage 112 and one or more non-volatile memory devices within non-persistent storage 112 include non-transitory computer-readable storage media. In some embodiments, non-persistent storage 111 or alternatively non-transitory computer-readable storage media sometimes store the following programs, modules, and data structures or a subset thereof in conjunction with persistent storage 112:

[0106] An optional operating system 116 (e.g., ANDROID, iOS, DARWIN, RTXC, LINUX, UNIX, OS X, WINDOWS, or an embedded operating system such as VxWorks), which includes procedures for handling various basic system services and for performing hardware-dependent tasks;

[0107] Optional network communication module (or instructions) 118 for connecting the system 100 to other devices and / or the communication network 104;

[0108] a dataset storage area 120 storing a plurality of datasets 122, each dataset comprising one or more identifiers (e.g., a sample identifier 124 and / or a cell / dataset identifier 126), an associated time period 128, and a cellular component vector 130 including one or more cellular components 132; and

[0109] A feature storage area 140 that stores one or more single-cell transition features 142 and one or more perturbation features 150 .

[0110] As described above, the dataset storage area 120 includes a plurality of datasets 120. Each dataset is a single cell measurement (e.g., Figure 3 The single cell measurement 310 of the cell is obtained (e.g., collected, communicated, etc.). The sample identifier (ID) 124 associated with each data set 122 indicates which sample the data set of the cell is from. The cell / data set identifier 126 indicates which cell and / or multiple data sets (e.g., a subset of data sets) the data set 122 is associated with and / or the state of the cell. In some embodiments, the time period 128 is associated with the capture time period of the data set 122 (e.g., when during cell growth, such as a first time period t0 when the cell is initially cultured, a second time period t1 when measurements of cell expression are taken, etc.).

[0111] In addition, in some embodiments, each data set 120 includes a cellular component vector 130, which includes one or more cellular components 132. In some embodiments, the one or more cellular components 132 include all cellular components of the cell or a subset of these cellular components of the cell. Each cellular component 132 represents a cell component associated with a measurement (e.g., Figure 3 The dimensions of the data associated with the single cell measurements 310 of the present invention are shown in Figure 1. Typically, the data sets 122 include a high (e.g., greater than 3, greater than 5, greater than 10, greater than 100, etc.) dimensionality, which includes a large amount of data. In addition, in some embodiments, each data set 122 is obtained from a cell in a plurality of cells (e.g., from a sample) that has transitioned from a "progenitor" cell type (e.g., from a first state to an altered state).

[0112] In some embodiments, the system includes a feature storage area 140 that stores one or more single-cell transition features 142 and one or more perturbation features 150. In some embodiments, the one or more single-cell transition features 142 include one or more predetermined features (e.g., training features). In some embodiments, the one or more single-cell transition features 142 include single-cell transition features determined by the system 100 and / or stored within the system for future use. Each single-cell transition feature 142 includes a cell component identification 144, and the cell component identification 144 also includes multiple cell components (e.g., Figure 1 In addition, each cellular component 132 associated with a single cell transition feature 142 includes a corresponding significance score 134. In some embodiments, dimensionality reduction (e.g., Figure 3 dimensionality reduction 320), which generates (e.g., stored in Figure 1 The reduced dimension component storage area 146-1 and / or the generated Figure 3 148 (e.g., Figure 1 Thus, in some embodiments, the system 100 performs dimensionality reduction (e.g., Figure 3 dimensionality reduction 320) to generate a plurality of reduced-dimensionality components 148 (e.g., to generate Figure 3 132 of the cellular components 132 of the data set 122, thereby preserving the underlying patterns present in the cellular components 132 of the data set 122. In some embodiments, the output of this dimensionality reduction (e.g., Figure 1 The dimensionally reduced components 148-1-1 through 148-1-F) of are matrices (eg, matrix M, referred to below) that encode the dataset 122 in a compressed form while also maintaining the underlying latent structure of the dataset.

[0113] In some embodiments, the feature transition store includes a manifold 149. In some embodiments, the manifold 149 is associated with a corresponding reduced dimension component 148 of a single-cell transition feature 142. The manifold 149 is identified by performing manifold learning with the cell component vectors 130 of the dataset 122 associated with the manifold (e.g., the dataset 122 associated with the single-cell transition feature 142).

[0114] The feature storage area 140 also includes one or more disturbance features 150 associated with the corresponding disturbance. Each disturbance feature includes a cellular component identification 152, which includes a plurality of cellular components (e.g., Figure 1 In some embodiments, the cellular components of the cellular component identification 152 include some or all of the cellular components associated with the corresponding single-cell transition feature 144 (e.g., the cellular component identification 152 of the perturbation feature 150-1 includes Figure 1 In addition, each cellular component of the perturbation feature 150 includes a corresponding significance score 134.

[0115] In various embodiments, one or more of the above identified elements are stored in one or more aforementioned memory devices, and correspond to the instruction set for performing the above functions. The above identified modules, data or programs (e.g., instruction sets) do not need to be implemented as separate software programs, processes, data sets or modules, and therefore the various subsets of these modules and data can be combined or otherwise rearranged in various embodiments. In some embodiments, non-persistent memory 111 optionally stores a subset of the above identified modules and data structures. In addition, in some embodiments, memory stores additional modules and data structures not described above. In some embodiments, one or more of the above identified elements are stored in a computer system, rather than in a computer system of system 100, and the computer system can be addressed by system 100 so that system 100 can retrieve all or part of these data when needed.

[0116] although Figure 1 A "system 100" is depicted, but the diagram is intended more as a functional description of various features that may be present in a computer system than as a schematic diagram of the structure of the implementations described herein. In practice, and as recognized by one of ordinary skill in the art, items shown separately may be combined and some items may be separated. Furthermore, although Figure 1 Certain data and modules are depicted in non-persistent memory 111, but some or all of these data and modules may alternatively be stored in persistent memory 112 or in more than one memory. For example, in some embodiments, at least dataset store 120 is stored in a remote storage device that may be part of a cloud-based infrastructure. In some embodiments, at least dataset store 120 is stored on a cloud-based infrastructure. In some embodiments, dataset store 120 and feature store 140 may also be stored in one or more remote storage devices.

[0117] Although referenced Figure 1 A system according to the present disclosure is disclosed, but now reference is made to Figure 2 The method 200 according to the present disclosure is described in detail.

[0118] Box 202. See Figure 2 At block 202, the method includes accessing (e.g., in electronic form) a single cell transition signature (e.g., Figure 1 The single cell transition signature 142-1 represents a measure of differential cellular component expression between a first cell state and an altered cell state. The altered cell state occurs by a cell transition from the first cell state to the altered cell state. The single cell transition signature 142 includes identification of multiple cellular components (e.g., Figure 1For each corresponding cellular component in the plurality of cellular components (e.g., Figure 1 For example, the first significance score 134-1-1 quantifies the association between the expression change of the corresponding cellular component and the change of the cellular state between the first cellular state and the changed cellular state.

[0119] In some embodiments, accessing the single-cell transition signature includes determining the single-cell transition signature 142. This determination is based on a first plurality of first single-cell cell component expression datasets (e.g., dataset 122-1, dataset 122-2, and dataset 122-3) and a second plurality of second single-cell cell component expression datasets (e.g., dataset 122-4, dataset 122-5, and dataset 122-6). Each corresponding first single-cell cell component expression dataset 122 in the first plurality of first single-cell cell component expression datasets is derived from a corresponding single cell of the first plurality of cells in the first cell state (e.g., Figure 3 In addition, each corresponding second single-cell cell component expression dataset in the second plurality of second single-cell cell component expression datasets is obtained from a corresponding single cell of the second plurality of cells in the changed cell state (e.g., Figure 3 Single cell measurements 310) were obtained.

[0120] In some embodiments, determining the single cell transition signature comprises determining a difference in the amount of the cellular component across the plurality of cellular components 132. The difference is a difference between a first plurality of first single cell cellular component expression data sets and a second plurality of second single cell cellular component expression data sets. In some embodiments, the difference is determined using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, or a generalized linear model.

[0121] In some embodiments, each respective dataset 122 of the first plurality of single-cell cellular component expression datasets includes a corresponding cellular component vector (e.g., Figure 1In addition, each corresponding data set of the second plurality of single cell cell component expression data sets includes a corresponding cell component vector in the second plurality of cell component vectors (e.g., cell component vector 130-2 of data set 122-2). Each corresponding cell component vector in the first plurality of cell component vectors and the second plurality of cell component vectors includes a plurality of elements. Each corresponding element in the corresponding cell component vector 130 is associated with a corresponding cell component 132 in the plurality of cell components and includes a corresponding value representing the amount of the corresponding cell component of the corresponding single cell, the corresponding single cell being represented by the corresponding data sets of the first plurality of single cell cell component expression data sets and the second plurality of single cell cell component expression data sets (e.g., cell components and values ​​of Table 2).

[0122] Additionally, in some embodiments, the cellular component 132 includes a plurality of genes. Additionally, in some embodiments, the one or more datasets 122 are generated using methods including single-cell RNA sequencing (scRNA-seq), scTag-seq, single-cell assay for transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination thereof (e.g., methods of Table 1).

[0123] Block 204. Referring to block 204, the method further includes accessing (eg, electronically) a disturbance signature (eg, Figure 1 The perturbation signature 150-1 represents a measure of differential cellular component expression between a plurality of unperturbed cells and a plurality of perturbed cells exposed to a perturbation. The perturbation signature 150 includes identification of all or a portion of the plurality of cellular components (e.g., Figure 1 For each corresponding cellular component (e.g., Figure 1 Cell components 132-3-1 to 132-3-D), corresponding to the second significant score (e.g., Figure 1 The significance score 134) quantifies the association between the expression changes of the corresponding cellular components between the multiple unperturbed cells and the multiple perturbed cells and the changes in the cell states between the multiple unperturbed cells and the multiple perturbed cells.

[0124] In some embodiments, method 200 includes performing dimensionality reduction (e.g., Figure 3 320). This dimensionality reduction generates multiple dimensionality reduction components (e.g., Figure 1148). In some embodiments, the dimensionality reduction is a principal component algorithm, a random projection algorithm, an independent component analysis algorithm or a feature selection method, a factor analysis algorithm, a Sammon map, a curvilinear components analysis, a stochastic neighbor embedding (SNE) algorithm, an Isomap algorithm, a maximum variance expansion algorithm, a local linear embedding algorithm, a t-SNE algorithm, a non-negative matrix factorization algorithm, a kernel principal component analysis algorithm, a graph-based kernel principal component analysis algorithm, a linear discriminant analysis algorithm, a generalized discriminant analysis algorithm, a unified manifold approximation and projection (UMAP) algorithm, a LargeVis algorithm, a Laplacian Eigenmap algorithm, or a Fisher's linear discriminant analysis algorithm. See, for example, Fodor, 2002, “A survey of dimension reduction techniques,” Center for Applied Scientific Computing, Lawrence Livermore National, Technical Report UCRL-ID-148494; Cunningham, 2007, “Dimension Reduction,” University College Dublin, Technical Report UCD-CSI-2007-7; Zahorian et al., 2011, “Nonlinear Dimensionality Reduction Methods for Use with AutomaticSpeech Recognition," Speech Technologies.doi:10.5772 / 16863.ISBN 978-953-307-996-7; and Lakshmi et al., 2016, "2016IEEE 6th International Conference on Advanced Computing (IACC),” pages 31–34.doi:10.1109 / IACC.2016.16, ISBN 978-1-4673-8286-1, each of which is incorporated herein by reference. Thus, in some embodiments, the dimensionality reduction is a principal component analysis (PCA) algorithm, and each corresponding extracted dimensionality reduction component includes a corresponding principal component derived by PCA. In such embodiments, the number of principal components in the plurality of principal components can be limited to a threshold number of principal components calculated by the PCA algorithm.The threshold number of principal component can be for example 5,10,20,50,100,1000,1500 or any other number.In some embodiments, each principal component calculated by the PCA algorithm is assigned eigenvalues ​​by the PCA algorithm, and the corresponding subset of the first plurality of extracted features is limited to the threshold number of principal components that assign the highest eigenvalues.For each corresponding cell component vector in the first plurality of cell component vectors and the second plurality of cell component vectors 130, the plurality of dimensionality reduction components are applied to the corresponding cell component vector to form a corresponding dimensionality reduction vector, the corresponding dimensionality reduction vector comprises the dimensionality reduction component value of each corresponding dimensionality reduction component in the plurality of dimensionality reduction components (for example, forming). Figure 3 This forms a corresponding first plurality of reduced dimensionality vectors and a second plurality of reduced dimensionality vectors. In addition, in some embodiments, the method includes performing clustering to generate a set of clusters Cj (e.g., Figure 3 Clusters 340 of clusters Cj are formed. Each cluster includes a plurality of points corresponding to a subset of the first plurality of reduced dimensionality vectors and the second plurality of reduced dimensionality vectors. A first plurality of cells from a first cluster of the set of clusters Cj and a second plurality of cells from a second cluster of the set of clusters Cj are both identified.

[0125] In some embodiments, method 200 includes performing manifold learning (e.g., Figure 3 Such manifold learning identifies the relative cell state of each cell in the first plurality of cells and the second plurality of cells relative to each other cell (e.g., generates Figure 3 For manifold learning, see, e.g., Wang et al., 2004, “Adaptive Manifold Learning,” Advances in Neural Information Processing Systems 17, which is incorporated herein by reference.

[0126] In some embodiments, the plurality of unperturbed cells are control cells (eg, cells not exposed to a perturbation). Additionally, in some embodiments, the unperturbed cells are an average of unrelated perturbed cells that have been exposed to a perturbation.

[0127] In some embodiments, the method includes pruning the single cell transition signature and / or perturbation signature. Such pruning limits the plurality of cellular components 132 (eg, limiting the cellular components to transcription factors).

[0128] In some embodiments, a measure of differential cellular component expression (e.g., Figure 3The method further comprises the step of quantifying the difference in the amount of cellular components between the third plurality of third single cell cellular component expression datasets and the fourth plurality of fourth single cell cellular component expression datasets (differentially expressed cellular components 350). Similarly, in some embodiments, the difference is determined using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, or a generalized linear model. In addition, each corresponding third single cell cellular component expression dataset 122 in the third plurality of third single cell cellular component expression datasets is obtained from a corresponding single cell in the plurality of unperturbed cells. In addition, each corresponding fourth single cell cellular component expression dataset in the fourth plurality of fourth single cell cellular component expression datasets is obtained from a corresponding single cell of a fourth plurality of cells in the plurality of perturbed cells exposed to a perturbation.

[0129] In some embodiments, for each respective cellular component in the plurality of cellular components, determining a corresponding second significance score for the respective cellular component comprises replacing the significance score for the respective cellular component with the corresponding matching score for the respective cellular component (e.g., replacing Figure 1 The significant score 134-dE of the cell component 132-1-1 is replaced with the significant score 134-1-1 associated with the cell component 132-1-1). In some embodiments, this replacement forms a matching score. The matching scores of the multiple cell components are combined to generate a matching score for the disturbance. Therefore, the matching score based on the corresponding disturbance determines whether the disturbance is associated with the transition of the cell between the first cell state and the changed cell state (e.g., whether it affects the cell transition). In some embodiments, the matching score includes a discrete score or a continuous score.

[0130] In some embodiments, replacing the score 134 includes replacing the significant score with a first score if both the cellular component amount 132 of the corresponding cellular component from the single-cell transition signature 142 and the cellular component amount 132 of the corresponding cellular component from the perturbation signature 150 are upregulated. Such replacement also includes replacing the significant score 132 with a second score if the cellular component amount from the single-cell transition signature 142 of the corresponding cellular component is upregulated and the cellular component amount from the perturbation signature 150 of the corresponding cellular component is downregulated. In addition, if the cellular component amount from the perturbation signature 150 of the corresponding cellular component is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0131] Block 206. Referring to block 206, method 200 includes comparing the single cell transition signature 142-1 to the perturbation signature 150-1. This comparison determines whether the perturbation will affect the cell transition.

[0132] In some embodiments, the method 200 includes filtering the single cell transition signature 142 and / or the perturbation signature 150. This filtering reduces the number of cellular components 132 included in the single cell transition signature 142 and the perturbation signature 150, which helps reduce the amount of data for the signature and facilitates performing the method 200 (e.g., performing Figure 3 The amount of time required for post-processing 360).

[0133] In some embodiments, method 200 includes comparing 206 (e.g., based on Figure 3 The perturbation is identified as a perturbation that promotes the changed cell state by post-processing 360. In some embodiments, different types of cell components are used to generate the single cell transition signature 142 and / or the perturbation signature 150. Similarly, in some embodiments, the same type of cell components are used to generate the single cell transition signature 142 and / or the perturbation signature 150.

[0134] II. Methods for Culturing Cells in Vitro to Perform Single Cell Analysis

[0135] In implementing the techniques described herein for identifying causes of cell fate, it is useful to generate data sets of measurements of cellular components obtained from single cells. To generate these data sets (e.g., via Figure 3 Single cell measurements 310 generated Figure 1 A cell population of interest is cultured in vitro. Single cell measurements of one or more cell components of interest 132 are performed at one or more time periods during the culture to generate a data set 122. (e.g., Figure 3 Single cell measurements 310). In some embodiments, the cell components of interest include: nucleic acids, including DNA, modified (e.g., methylated) DNA; RNA, including coding (e.g., mRNA) or non-coding (e.g., sncRNA) RNA; proteins, including post-transcriptionally modified proteins (e.g., phosphorylated, glycosylated, myristylated, etc.); lipids; carbohydrates; nucleotides (e.g., adenosine triphosphate (ATP), adenosine diphosphate (ADP), and adenosine monophosphate (AMP)), including cyclic nucleotides, such as cyclic adenosine monophosphate (cAMP) and cyclic guanosine monophosphate (cGMP); other small molecule cell components, such as oxidized and reduced forms of nicotinamide adenine dinucleotide (NADP / NADPH); and any combination thereof. In some embodiments, cell component measurements include gene expression measurements, such as RNA levels.

[0136] Any of a number of single cell cellular component expression measurement techniques can be used to collect dataset 122 (e.g., techniques of Table 1, Figure 1The single cell measurement 310 of the technology, etc.). Examples include, but are not limited to, single cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single cell assay of transposase accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, etc. The cell component expression measurement can be selected based on the desired cell component to be measured. For example, scRNA-seq, scTag-seq, and miRNA-seq measure RNA expression. Specifically, scRNA-seq measures the expression of RNA transcripts, scTag-seq enables detection of rare mRNA species, and miRNA-seq measures the expression of microRNAs. CyTOF / SCoP and E-MS / Abseq measure protein expression in cells. CITE-seq simultaneously measures both gene expression and protein expression in cells, and scATAC-seq measures chromatin conformation in cells. Table 1 below provides links to exemplary protocols for performing each of the above-mentioned single cell cell component expression measurement technologies.

[0137] Table 1 - Exemplary measurement scheme

[0138]

[0139]

[0140] The cell component expression measurement technique used may result in cell death. Alternatively, the cell component can be measured by extraction from living cells, for example by extracting the cytoplasm without killing the cell. This diversity of techniques allows the same cell to be measured at multiple different time points.

[0141] If the cell population is heterogeneous, such that multiple different cell types derived from the same "progenitor" cell are present in the population, then single cell cell component expression measurements can be performed at a single time point or relatively few time points while the cells are grown in culture. Due to the heterogeneity of the cell population, the collected data set 122 will represent various types of cells along the transformation trajectory.

[0142] If the cell population is substantially homogeneous, such that only a single or relatively few cell types, primarily "progenitor" cells of interest, are present in the population, then single cell cellular component expression measurements can be performed multiple times over a period of cellular transformation.

[0143] For each cell and at each time period (e.g., Figure 1The single cell cell component expression data set 122 is generated when the time period 128) is applicable. The single cell cell component expression measurement results collected from the cell group at multiple different time points can be collectively interpreted as a "pseudo-time" representation of cell expression over time of the cell type derived from the same "ancestor" cell. The term pseudo-time is used in two aspects: first, the cell state transitions between cells are not necessarily the same, and therefore the cell group provides a distribution of the transition process that the cells of the "ancestor" type may experience over time; and second, the cell component expression measurements expressed by those multiple cells at multiple time points simulate possible transition behaviors over time, even if the cell component expression measurements of different cells generate data sets. As an example of intentional simplification, even if cell X gives a data set at time point A, and cell Y gives a data set at time point B, these two data sets together represent the pseudo-time of the transition between time point A and time point B.

[0144] For ease of description, two such data sets 122 captured for the "same" cell at two different time periods (e.g., first time period 128-1 for first data set 122-1, second time period 128-2 for second data set 122-2, etc.) (assuming the use of techniques that do not kill the cell as described above) are referred to herein as different "cells" (and corresponding different data sets), because in practice such cells will often be slightly or significantly transformed from each other, in some cases having completely different cell types as determined from the relative amounts of various cellular components. From the context, these two measurements of a single cell at different time points can be interpreted as different cells for analysis purposes because the cell itself has changed.

[0145] Note that the methods described herein for expressing cells (e.g., Figure 1 Cell / dataset identifier 126) / time period (e.g., Figure 1 The separation of data sets by time period 128) is for clarity of description; in practice, these data sets can be stored in a computer memory and can be logically operated on as one or more aggregated data sets (e.g., by cell in all time periods, for all cells and time periods at once).

[0146] In some embodiments, it is useful to collect a data set 122 in which the "progenitor" cells of interest have been disturbed from their baseline state. There are many possible reasons for doing this, such as knocking out (e.g., removing, abolishing, etc.) one or more cellular components, evaluating the difference between a healthy cell state and a diseased cell state, etc. In these embodiments, the method may also include a step of introducing the desired modification to the cell. For example, one or more disturbances may be introduced into the cell, a custom virus designed to knock out one or more cellular components may be introduced, CRISPR may be used to edit cellular components, etc. Examples of techniques that can be used include, but are not limited to, RNA interference (RNAi), transcription activator-like effector nucleases (TALENs), or zinc finger nucleases (ZFNs).

[0147] Depending on the mode of applying the disturbance, not all cells will be disturbed in the same way. For example, if a virus is introduced to knock out a specific gene, the virus may not affect all cells in the population. More generally, the property can be advantageously used to evaluate the impact of many different disturbances on a single population. For example, a large number of customized viruses can be introduced, each of which performs different disturbances, such as causing different genes to be knocked out. The virus will infect some subsets of various cells differently, knocking out the gene of interest. Then, single cell sequencing or another technology can be used to identify which viruses affect which cells. The resulting different single cell sequencing data sets can then be evaluated to identify the impact of gene knockout on gene expression according to the method described elsewhere in this specification.

[0148] Other types of multi-perturbation cell modifications can be performed similarly, such as introduction of multiple different perturbations, barcoded CRISPR, etc. In addition, more than one type of perturbation can be introduced into the cell population to be analyzed. For example, cells may be subjected to different influences (e.g., different viruses introduced), and different perturbations may be introduced into different cell subpopulations.

[0149] In addition, different subsets of a cell population can be perturbed in different ways, rather than simply mixing many perturbations and evaluating after the fact which cells are affected by which perturbations. For example, if a cell population is physically separated into different wells of a multi-well plate, a different perturbation can be applied to each well. Other ways of achieving different perturbations for different cells are also possible.

[0150] Below, single cell gene expression measurement is used to illustrate the method. It should be understood that this is illustrative rather than restrictive, because the present invention encompasses similar methods for the measurement of other cellular components obtained from single cells. It should also be understood that the present invention encompasses methods for the measurement obtained using experimental work performed directly by individuals or organizations practicing the methods described in the present disclosure, and methods for the measurement obtained using indirect reports of the results of experimental work performed by others and obtained by any means or mechanism (including third-party publications, databases, assays performed by contractors, or other sources of suitable input data that can be used to practice the disclosed methods).

[0151] As discussed herein, gene expression in a cell can be measured by sequencing the cell and then counting the amount of each gene transcript identified during sequencing. In some embodiments, the gene transcripts sequenced and quantified can include RNA, such as mRNA. In alternative embodiments, the gene transcripts sequenced and quantified can include downstream products of mRNA, such as proteins, such as transcription factors. Generally, as used herein, the term "gene transcript" can be used to represent any downstream product of gene transcription or translation (including post-translational modifications), and "gene expression" can be used to refer generally to any measure of gene transcripts.

[0152] Although the remainder of this specification focuses on the analysis of gene transcripts and gene expression, all techniques described herein are equally applicable to any technique that obtains data about those cells on a single cell basis. Examples include single cell proteomics (protein expression), chromatin conformation (chromatin state), methylation or other quantifiable epigenetic effects.

[0153] The following description provides methods for culturing a cell population in vitro to culture a cell population for various time periods (e.g., Figure 1 Single cell cell component expression measurements are performed over multiple time periods 128) (e.g., Figure 3 An exemplary general description of the measurement 310 of . In general, methods for culturing cells in vitro are known to those skilled in the art. Those skilled in the art will also understand how to modify the methods to grow for longer or shorter periods of time, perform additional or fewer single cell measurement steps, etc.

[0154] In one embodiment, a method for culturing a cell in a first cell state into a cell in an altered cell state comprises one or more of the following steps:

[0155] Day 0: Thaw a number of cells in the first cell state onto a plate in a medium appropriate for cell growth.

[0156] Day 1: A certain number of cells in the first cell state are seeded into a multi-well plate. If applicable, additional steps are performed to affect the cellular components of the cells. For example, co-infection with one or more viruses to knock out the cellular components of interest.

[0157] Perform cellular component expression measurement iteration t1 on the cells in the well.

[0158] Day 1+1: If any additional treatments are performed, change the media as needed.

[0159] If applicable, perform cell component expression measurements for the cells in the well for iteration t l .

[0160] Day 1+m: Change the culture medium to one appropriate to support the growth of cells in the altered cell state.

[0161] If applicable, perform cell component expression measurements for the cells in the well for iteration t m .

[0162] Day 1+n, o, p, etc.: Change media as needed to support further cell state transition from the first cell state to the altered cell state. If applicable, perform additional steps to achieve further transition from the first cell state to the altered cell state. For example, add a perturbation of interest to push cells toward the altered cell state.

[0163] If applicable, perform cell component expression measurements for the cells in the well for iteration t n ,t o ,t p wait.

[0164] Day q: Perform cell component expression measurements for cells in the wells in altered cell states Iteration t q .

[0165] The plate is fixed and stained with antibodies that match the cell components / proteins of interest to sort / identify cells without lysis / destruction of these cells to be measured. It can also be used to identify surface markers that may not be seen with the same high resolution in the cytoplasmic environment. By scanning each hole with a cell imaging system such as Molecular Devices HCIIXM4 imaging. Quantify the number of cells in the desired altered cell state in each hole.

[0166] Table 2 shows a fragment of multiple data sets 122, including single cell expression measurements that can be obtained from a cell population at one or more time points (e.g., Figure 3 The Sample ID column indicates which sample the cell's data came from (e.g., Figure 1 In practice, cells in a population may be taken from more than one sample (e.g., first sample identifier 124-1, second sample identifier 124-2, etc.), each of which may be from the same or different subjects. The cell or dataset column indicates which cell or dataset the data for a given row is associated with (e.g., Figure 1 The data set 122 may alternatively be represented as a vector r of data. i (For example, Figure 1 The Time Period column indicates when during the growth of the cell the data set for that row was captured (if relevant) (e.g., Figure 1 time period 128-1).

[0167] The remaining columns of Table 2 correspond to the cellular components of interest (cellular components 132-1-1 to 132-1-B) of the cell. This can be all of the cellular components of the cell, or just a subset. Each cellular component 132 is associated with a different column. If the data set is represented as a vector r i , then each cellular component corresponds to entry i in the vector. In some embodiments, the value for each cell may be an (integer) count of the number of cellular components as measured by single-cell expression, or some normalized (rational) form thereof.

[0168] Table 2 - Exemplary Datasets

[0169]

[0170] III. Methods for analyzing single-cell datasets to determine differential expression of cellular components

[0171] III.A. Overview

[0172] A cell state transition (i.e., a transition of a cell state from a first cell state to an altered cell state) is marked by a change in the expression of a cellular component 132 in the cell. For example, a transition can be marked by a change in the expression of a cellular component 132 in the cell, and therefore by the identity and amount of the cellular component (e.g., mRNA, transcription factor) produced by the cell. However, at least currently, due to the complexity of intracellular activity, cell state transitions are not completely deterministic. In an attempt to gain insight into this complexity, this description applies statistical techniques to a single cell dataset 122 that quantifies cellular components 132 in cells of a cell population under the following theory: Different cellular component expressions associated with different presence, absence, or amounts of one or more measured cellular components of interest at different stages of a cell state transition provide a high dimensional dataset (e.g., Figure 1Here, the high dimensionality of the data originates from each cell component measurement contained in the data set 122. Each cell component 132 represents a dimension, and the cell component measurement data set 122 of each cell component can have a shape that encodes potential information about the biological process of the "progenitor" cell transitioning to a different cell type. In practice, the number of cell components 132 can be about thousands to tens of thousands, making the calculations described herein impractical, if not impossible, to perform mentally or manually.

[0173] In general, these statistical techniques can be characterized as methods in which high-dimensional data are compressed into a lower-dimensional space while preserving the shape of any underlying information encoded in the data set (e.g., Figure 1 The cell component vector 130 is reduced to 320 Figure 3 Matrix M). Evaluate low-dimensional data to identify the cellular components that exist between the different stages of cell state transition. Because the input data of the method is a single cell cellular component expression data set 122 of a plurality of cellular components of interest on a per cell basis, the set of differentially expressed cellular components therefore represents which cellular components have statistically significant over-representation or under-representation in terms of presence, absence or amount relative to other cellular components of the cell. Any of many methods and metrics can be used to identify which of those cellular components are fully "difference" expressed relative to other cellular components, so as to be labeled as "differentially expressed" according to this description. Since the cell population that can obtain data set 122 includes cells of different types and different transition stages, knowing which cellular component differences exist (for example, which cellular component differentially expresses) provides which cellular component influences the expression of active cellular components in the process of transition or other transitions or related insights thereto.

[0174] III.B. Use Cases

[0175] No matter which type of method is used, the determination of the cellular component of differential expression can change according to the result sought.For example, if used method is identified as in pedigree or outside pedigree by specific cell, then the determination of which cellular component differential expression can be performed by comparing the expression level of the cellular component of the cell in the pedigree and the expression level of the cellular component of the cell outside the pedigree.The relative expression of these cellular components indicates which cellular components are active in the cell of a type or another type individually or in combination.As mentioned above, described expression data can be used to identify the cellular component subset to be marked as differential expression.Then can be determined causality by knocking out the cellular component of identification in vitro and evaluating whether the cell fate of the experimental cell group is subject to the active variation influence of cellular component.

[0176] As another example, if the method used identifies a particular cell as within a lineage, and identifies other cells as "progenitor" cells or intermediate cells along a transition trajectory to a cell type within a lineage, the determination of which cell components are differentially expressed can be performed by comparing the expression levels of cell components of cells determined as within the lineage with the expression levels of cell components of cells determined as "progenitor" cells and / or intermediate cells of cells within the lineage. As described in the previous paragraph, the relative expression of these cell components indicates which cell components are active in one type or another type of cell, either alone or in combination, and the expression data can be used again to identify a subset of cell components to be marked as differentially expressed. Also as described above, causality can then be determined by knocking out the identified cell components in vitro and evaluating whether the cell fate of the experimental cell population is affected by changes in the activity of the cell components.

[0177] As another example, a cell population may include two subpopulations of cells, a healthy subpopulation and an unhealthy subpopulation. During cell culture, a variety of different perturbations may be introduced into the unhealthy subpopulation. By combining subsequent single cell expression measurements with the methods described herein, it may be determined what effect the perturbations have on the differential cellular component expression of cellular components in the unhealthy subpopulation, particularly with respect to the healthy subpopulation. For example, a subset of cells from an unhealthy subpopulation exposed to one or more perturbations may exhibit cellular component expression consistent with a healthy cell subpopulation, indicating that the perturbations have a desired effect on the unhealthy cell subpopulation.

[0178] III.C. Using low-dimensional data to identify differentially expressed cellular components

[0179] Figure 3 is a flow chart of a first example of a differential cellular component expression assay for determining a set of differentially expressed cellular components 132 according to one embodiment. Figure 3 A non-limiting, illustrative embodiment of the general case of using differential cell component expression descriptions is provided. At step 310, single cell expression measurements are performed to generate a plurality of data sets 122 for a cell population, as discussed above in Section II. As described above, each data set 122 for each cell can be represented as a vector r of cell components. i (For example, Figure 1 130), which includes l kinds of cellular components (e.g., Figure 1 The data set 122 obtained from the single cell expression measurement 310 is typically stored in a digital format on a computing device (e.g., Figure 1 of the system 100) (e.g., Figure 1 112), however, they can be loaded into active storage (e.g., Figure 1The non-persistent memory 111) is used to perform the remaining steps described herein. Figure 3 The remaining steps of the method are performed by one or more computing devices (e.g., Figure 1 The system 100 is performed. Figure 1 Discussing exemplary computing devices, however, in practice, Figure 3 The methods may include additional fill-in or subsequent steps that may be performed outside of the computer, such as additional in vitro testing or clinical decisions based on the results of the steps described herein.

[0180] III.C.1. Dimensionality reduction

[0181] As described above, since each cell component 132 represents data of different dimensions, the data set 122 has a high dimensionality overall. In step 320, a dimensionality reduction is performed by a computing device (e.g., system 100) to reduce the dimensionality of the data while maintaining the structure of any potential patterns present in the cell component 132 amounts of the data set 122.

[0182] The input to the dimensionality reduction step 320 is typically a matrix, similar to Table 2 above, which connects the expression vectors of individual cells (e.g., Figure 1 The output of dimensionality reduction 320 is a matrix, referred to herein as matrix "M" for simplicity, which encodes the original data in a compressed form while maintaining the underlying potential structure of the data. Each row in the matrix M is associated with a specific one in the cell. Each column in the matrix M is associated with a dimension in the dimensionality reduction space provided by the dimensionality reduction. The values ​​in the entries at each row-column grouping are determined by dimensionality reduction based on the original input data set.

[0183] In some embodiments, these dimensionality reduction techniques result in some lossy compression of the data, however, the resulting output matrix M is smaller in computational storage size and therefore requires less computational processing power to analyze along with the other downstream techniques discussed in the remaining steps of the method, making it computationally feasible to obtain the results of those steps in a reasonable time with current generation computing devices.

[0184] A variety of dimensionality reduction techniques can be used. Examples include, but are not limited to, principal component analysis (PCA), non-negative matrix factorization (NMF), linear discriminant analysis (LDA), diffusion mapping or (neural) network techniques such as autoencoders.

[0185] Each of the techniques mentioned in these paragraphs operates differently to extract the main drivers of variation and reduce the dimensionality of the original input data, but each technique outputs a matrix M in a lower dimensional space.

[0186] III.C.2. Manifold Learning

[0187] The reduced dimensional data in the matrix M (e.g., the reduced dimensionality component storage area 146) is significantly reduced in dimensionality relative to the original high dimensional data from the single cell expression dataset 122. However, the resulting matrix M is embedded in a nonlinear manifold (e.g., Figure 1 In step 330, a manifold learning technique is applied to the matrix M to extract the manifold. The manifold 149 itself not only provides useful information about differential cellular component expression between cells in pseudo-time, but it can also be used to visualize the information.

[0188] The input to the manifold learning step 330 is the matrix M from the dimensionality reduction step 320. The output of the manifold learning 330 is another matrix, referred to herein as the matrix "N" or the manifold (e.g., Figure 1 149). The structure of matrix N is such that each row of matrix N corresponds to one of the original cells of the population, referred to herein as a "point" for the remaining steps of the method. In one embodiment, matrix N has two columns, arbitrarily referred to as the X dimension and the Y dimension, corresponding to the two dimensions that the manifold learning step 330 is configured to output, regardless of the particular manifold learning algorithm used. The X and Y dimensions are determined by the manifold learning step, and which dimension is selected to best fit the data from matrix M depends on which manifold algorithm is used. Figure 4B As shown in , a manifold with two such columns facilitates visualization. In other embodiments, the manifold matrix N has additional dimensions beyond the two-dimensional form presented herein.

[0189] An exemplary matrix N is provided in Table 3 below. Figure 4B A plot of data from Example 1 below is provided in an embodiment using a force-directed layout in the dimensionality reduction step. Figure 4B The plots in are examples of results obtained according to the described methods, because in the described and similar exemplary experiments, the points are separated in the X / Y dimension along one or more tracks in the X / Y plane, where typically "progenitor" cells appear in one general region in X / Y space, spread toward intermediate cells in another general region in X / Y space, and end up at one or more different regions in X / Y space, which in practice are typically verified as cells that have transformed within the lineage or cells that have transformed outside the lineage. Typically, the number of regions and tracks identified depends on the type of "progenitor" cell and the type of cell into which the "progenitor" cell is known to transform. In addition, the regions of the points typically have a certain amount of spread between them, indicating that the cells are at different stages of progression in the transformation process.

[0190] Table 3 - Output matrix N

[0191]

[0192] A variety of rheological learning techniques can be applied to the matrix M to generate the matrix N. Examples include, but are not limited to, force-directed layout (Fruchterman, TM and Reingold, EM (1991). Graphics are drawn using force-directed layout. Software: Practice and experience, 21 (11), 1129-1164) (e.g., Force Atlas 2), t-distributed random neighbor embedding (t-SNE), local linear embedding (Roweis, ST and Saul, LK (2000). Nonlinear dimensionality reduction is performed using local linear embedding. Science, 290 (5500), 2323-2326 ) , Local Linear Isometric Mapping (ISOMAP, Tenenbaum, JB, DeSilva, V. and Langford, JC (2000). A global geometric framework for nonlinear dimensionality reduction. Science, 290 (5500), 2319-2323), kernel PCA, graph-based kernel PCA, affinity-based trajectory embedding thermal diffusion potential (PHATE), generalized discriminant analysis (GDA), unified manifold approximation and projection (UMAP) or kernel discriminant analysis. Discriminant analysis can be used, especially when some information about the specific cell type of each cell is known in advance. Force-directed layouts are useful in various specific embodiments because they are able to identify new, lower dimensions that encode nonlinear aspects of the underlying data caused by underlying biological processes such as cell state transitions. Force-directed layouts use a physics-based model as a mechanism for determining the reduced dimensionality that best represents the data. As an example, force-directed layout uses a form of physics simulation, where, in the described embodiment, each cell / data set in the collection is assigned a "repulsive" force, and there is a global "gravitational force" that, when calculated over all cells, identifies sectors of data that "spread" together under these competing "forces." Force-directed layout makes few assumptions about the structure of the data, and does not impose denoising methods.

[0193] Note that performing manifold learning 330 is an optional step. In some embodiments, manifold learning is not performed.

[0194] III.C.3. Clustering

[0195] In step 340, clustering is performed to generate j clusters C. j, to identify patterns of locations of points in the low-dimensional space provided by dimensionality reduction 320 (e.g., corresponding to subsets of the associated plurality of reduced dimensionality vectors 146). These clusters are used to cluster similar points (cells / datasets) to extract statistically relevant information about groups of points that are similar to each other in the low-dimensional space (e.g., a first cluster, a second cluster, etc.). Table 4 below shows exemplary clusters of points that may be output by clustering 340.

[0196] Table 4 - Cluster assignment

[0197]

[0198] Any of a number of clustering techniques can be used, examples of which include, but are not limited to, hierarchical clustering, k-means clustering, and density-based clustering. In a specific embodiment, a hierarchical density-based clustering algorithm (called HDBSCAN, Campello, RJ, Moulavi, D., Zimek, A. and Sander, J. (2015)) is used. Hierarchical density estimation for data clustering, visualization, and outlier detection. ACM Transactions on Knowledge Discovery from Data (TKDD), 10 (1), 5). In another embodiment, a clustering algorithm based on community detection is used, such as Louvain clustering (Blondel, VD, Guillaume, JL, Lambiotte, R. and Lefebvre, E. (2008). Rapid discovery of communities in large networks. Journal of statistical mechanics: theory and experiment, 2008 (10), P10008).

[0199] For clustering, these techniques use the data of a matrix M to determine the clusters. Regardless of the algorithm, in general, points that are closer to each other in the multidimensional space of the matrix M are more likely to be assigned to the same cluster, while points that are further away from each other are less likely to be assigned to the same cluster. Fig. 7A Provided by Figure 4B A plot of exemplary data of , where cluster assignments 1-10 are indicated with different visual markers for each point. The number of clusters may be set or constrained by the operator and / or determined dynamically based on the algorithm used.

[0200] III.C.4. Determination of differential cellular component expression

[0201] Dimensionality reduction 320, optional manifold learning 330, and clustering 340 steps are typically used to organize the cells of the population and their corresponding single-cell expression data sets 122 into clusters within a reduced dimensionality space, so that the underlying per-cell component expression measurement data can be aggregated and analyzed to extract meaningful information. In some embodiments, this reduced dimensionality space further reduces the amount of time and / or processing power required to complete the methods of the present disclosure.

[0202] One piece of information that can be obtained from clustering is which cellular components are differentially expressed in the population relative to other cells. In this paper, this set of cellular components is referred to as differentially expressed cellular components E. k The set, such as Figure 3 350. Some exemplary use cases for generating sets of differentially expressed cellular components are discussed above in Section III.B.

[0203] There are many clustering methods that use C j In one embodiment, determining whether a given cellular component (e.g., cellular component A) is differentially expressed is performed by evaluating the amount of cellular component A in a point (cell) in a given cluster C1 relative to one or more other clusters C m The amount of cellular component A at the points in m is determined, where m is not equal to 1. Standardization can also be used. For example, the expression level of the cellular component in the cell as a whole may vary from cell to cell for reasons that are independent of the cell state transition biology. Therefore, the cellular component amount can be standardized based on the total amount of the cellular component amount of each cell in the data set.

[0204] As discussed in Section III.B above, which cluster of cellular component A is compared to a given cluster C1 can vary depending on the implementation. Other clusters used for comparison can be clusters most strongly associated with cell types within the lineage, clusters most strongly associated with cell types outside the lineage, clusters most strongly associated with "progenitor" cell types, clusters most strongly associated with intermediate cell types, etc. Comparisons can also be made for more than one other cluster.

[0205] Given the comparison, cellular component A can be identified as differentially expressed based on any of a number of metrics, such as the total cellular component amount for each cluster (again, for all points in the cluster, or some aggregated metric, such as the mean, etc.), the normalized cellular component amount for each cluster, the median, mean, or other aggregated cellular component amount for each cluster, the ratio of the cellular component amount expressed relative to other cellular components, etc. In one embodiment, the criterion for establishing that cellular component A is differentially expressed is a threshold requirement.

[0206] For example, the normalized cellular component amount of cellular component A in cluster C1 may exceed the normalized cellular component amount of cellular component A in one or more other clusters Cm by at least a threshold value.

[0207] The determination of the cellular component of differential expression can also be relative.In one embodiment, the distance metric or other similar metrics of the standardized cellular component amount of multiple cellular component / clustering combinations, multiple cellular component / clustering combinations can be calculated.These metrics can be sorted according to sorting criteria (for example, the standardized cellular component amount highest in the cluster), and the cellular component or cellular component / clustering combination in the front of the sorting can be determined as the cellular component of differential expression.

[0208] In one embodiment, the amount of a given cell component in a given cluster can be used to identify which cell components are differentially expressed. In one embodiment, these differentially expressed cell components are identified using one of a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model.

[0209] Those skilled in the art will appreciate that other metrics involving the amount of cellular components per cellular component / cluster combination are possible.

[0210] III.C.5. Post-processing

[0211] Differentially expressed cellular components E k The set itself represents a useful output. However, it may be useful to further analyze the set of 360 differentially expressed cellular components to identify subsets of the set.

[0212] In one embodiment, a transcription factor database (e.g., Figure 1 The feature storage area 140 of the present invention can be used to screen the set of differentially expressed cellular components to identify the set of transcription factors associated with the cellular components present in the set. As an example, the information can be obtained from a ChIP-seq data set (information about which transcription factors bind to which regions of DNA, which is compared with the cellular components).

[0213] The dataset 122 discussed herein for a particular cell, such as the original input dataset r (e.g., Figure 1 Dataset 122-1) or differently expressed cellular components E k The set of and corresponding data sets may be missing cell component amounts due to many reasons (e.g., technical noise, dropouts, low cell component amounts, etc.). Taking these and any additional confounding factors into account, a simple model can be fit to the data set.

[0214] III.D. Prediction of perturbations affecting cell state transitions

[0215] By matching the differential cell component expression characterizing the specific cell transformation with the differential cell component expression caused by the cell exposure to disturbance, the disturbance affecting the specific cell state transformation can be predicted.Cell disturbance includes any treatment of cells with one or more compounds.The one or more compounds may include, for example, small molecules, biological preparations, proteins, proteins combined with small molecules, ADC, nucleic acids (such as siRNA or interfering RNA), cDNAs overexpressing wild-type and / or mutant shRNA, cDNAs overexpressing wild-type and / or mutant guide RNAs (e.g., Cas9 systems or other cell component editing systems) or any of the aforementioned any combinations.The differentially expressed cell components of specific cell transformations can be compared with the differentially expressed cell components caused by the cell exposure to disturbance.Then, the disturbance of the differential cell component expression causing the differential cell component expression of matching specific cell transformations can be predicted to affect specific cell transformations.

[0216] In order to predict the perturbation that affects a specific cell transition by matching the differential cell component expression that characterizes the specific cell transition with the differential cell component expression caused by exposure of the cell to the perturbation, first, identify the cell components that characterize the specific cell transition with the greatest expression difference. In some embodiments, these differentially expressed cell components are identified using one of a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model. In alternative embodiments, any statistical method can be used to identify the cell components that are most differentially expressed for a specific cell transition. The resulting ranked table (or list) of cell component 132 names and significance scores 134 can also be referred to as a 'single cell transition signature' (e.g., including Figure 1 The single cell transition signature 142 of each cellular component 132 is quantified by a significant score 134 for each cellular component 132. The association between the change in cellular component expression of the cellular component and the change in cell type between the original cell type and the cell type that is being transformed. In summary, these scores 134 form an overall measure of the differential cellular component expression associated with the transition between the original cell type (first cell state) and the cell type that is being transformed (the altered cell state).

[0217] Similarly, differential cellular component expression resulting from exposure of cells to a perturbation is identified for one or more perturbations. In some embodiments, to identify differential cellular component expression resulting from exposure of cells to a perturbation, the cellular component expression in cells exposed to the perturbation is compared to the cellular component expression in one or more control cells that have not been exposed to the perturbation or to the average of an unrelated perturbed sample (e.g., Figure 3The comparison is performed using a post-processing 360 of the present invention. In some embodiments, the comparison is performed using a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model. In alternative embodiments, any statistical method can be used to perform the comparison. In further alternative embodiments, the differential cell component expression caused by cell exposure to a disturbance can be known and identified from the literature. The resulting similarly ranked table (or list) of cell component names and significant scores can be referred to as a 'disturbance feature'.

[0218] In some embodiments, to reduce confounding due to technical variation, different experimental assays, and other variables in identifying single cell transition signatures and perturbation signatures, one or both of the signatures are filtered to include only transcription factors, which are proteins known to drive expression of certain cellular components. These transcription factors can be identified, for example, from the literature.

[0219] In some embodiments, to further reduce confusion due to technical variations and ambiguity of cell transitions, the cell components with the greatest expression differences of one or both features are truncated (or filtered or subsetted) at a given p-value and / or at a threshold number of cell components. The resulting truncated set of differentially expressed cell components exposed to cell transitions and perturbations is unordered and may contain between 10 and 25 cell components, or more or less, depending on the embodiment.

[0220] In the recognition and any processing of one or two features (e.g. Figure 1 After the single cell transition signature 142 and / or perturbation signature 150 is generated, the differentially expressed cellular components of the single cell transition signature 142 are compared with the differentially expressed cellular components of the perturbation signature 150. In one embodiment, to perform the comparison, the differentially expressed cellular components of the perturbation are represented as a matrix (e.g., Figure 3 The matrix M, Figure 1 130, etc.). Each row of the matrix is ​​associated with a single perturbation. Each column on the matrix is ​​associated with one of the differentially expressed cellular components. Each entry in the matrix includes a significance score 134 (e.g., p-value, t-score) for the differentially expressed cellular component 132 identified for a particular perturbation. The matrix is ​​a subset that includes only the differentially expressed cellular components identified for the single cell transition signature 142. This filtering can be done using the methods described in the previous paragraph (e.g., by a threshold p-value, by a threshold number of cellular components, etc.).

[0221] Each significant score 134 in the matrix is ​​replaced with a discrete matching score. In order to replace each significant score with a discrete matching score, the significantly upregulated cellular components 132 of the cell transition and the significantly downregulated cellular components of the cell transition are identified. For each of the significantly upregulated cellular components identified by the single cell transition marker 142, if the cellular component is also significantly upregulated for the disturbance feature 150 of the disturbance, the significant score in the matrix of the cell component / disturbance combination is replaced with a discrete matching score of '1'. If the cellular component of the disturbance feature is significantly downregulated relative to the single cell transition feature, the significant score in the matrix of the cell component / disturbance combination is replaced with a discrete matching score of '-2'. If the cellular component of the disturbance feature is not significantly upregulated or downregulated, the significant score in the matrix of the cell component / disturbance combination is replaced with a discrete matching score of '0'.

[0222] Conversely, for each of the significantly downregulated cellular components identified in the single cell transition signature, if the perturbed cellular component is also significantly downregulated, the significant score in the matrix of the cellular component / perturbation combination is replaced with a discrete matching score of '-1'. If the perturbed cellular component is significantly upregulated, the significant score in the matrix of the cellular component / perturbation combination is replaced with a discrete matching score of '2'. If the perturbed cellular component is not significantly upregulated or downregulated, the significant score in the matrix of the cellular component / perturbation combination is replaced with a discrete matching score of '0'. Those skilled in the art will appreciate that in some embodiments, these specific score replacements can be replaced with other values.

[0223] The result is a matrix where the number of rows is given by the number of perturbations and the number of columns is given by the differential cellular components from the single-cell transformations, and the entries represent the matching scores described above.

[0224] After replacing the significant scores in the matrix with discrete matching scores as described above, the discrete matching scores in each row of the matrix are summed to generate a total matching score for each row. The rows of the matrix are then sorted in descending order of the total matching scores, with each row corresponding to a perturbation. The top-ranked rows are associated with the perturbations most likely associated with the identified cell transitions of the single cell transition signature.

[0225] In some embodiments, for the total matching score of each row in the matrix, the estimation of the wrong cell component discovery rate is estimated. In order to estimate the wrong cell component discovery rate, the empirical marginal expression frequency of each cell component is calculated, and the empirical marginal expression frequency of each cell component is summed up on their combination, which generates the probability of accidentally identifying a given number of cell components (observing the probability of at least the same rare expression as seen in the data set for generating features), assuming independent distribution expression. Then, the probability can be used to calculate the wrong cell component discovery rate.

[0226] In certain embodiments, there may be covariates of disturbances. For example, if the disturbance is a small molecule, the covariates of the small molecule may include a specific dose of the small molecule, a time to measure cells exposed to the small molecule to quantify cellular components, and / or the identity of the cells exposed to the small molecule (e.g., a cell line). In some embodiments, the disturbance is predicted to affect a specific cell transition only when a threshold amount of the covariates of the disturbance is also predicted to affect a specific cell transition. For example, the disturbance can be predicted to affect a specific cell transition only when at least two of the covariates of the disturbance are also predicted to affect a specific cell transition.

[0227] Alternative matching methods may be used. For example, cell components may be matched to a database using a web interface (e.g., such as L1000CDS2, an ultrafast LINCS L1000 feature direction feature search engine, available on the World Wide Web at amp.pharm.mssm.edu / L1000CDS2 / # / index). This matching method does not perform as well as the matching method described in the previous paragraph, which produces results with much higher sensitivity, scales better, covers much more data (millions of samples instead of tens of thousands), takes into account significant overlap, disregards significant inconsistencies, and ignores non-significant information in the features.

[0228] Because the expression of cellular components of a particular single-cell state transition is highly variable and because the expression of cellular components affected by the perturbation is highly variable, it may be difficult to find a perturbation that matches a particular single-cell state transition. To alleviate this problem, in some alternative embodiments, the matching and subsequent identification of perturbations that affect cell state transitions along a particular trajectory can be performed by a trained neural network model.

[0229] Examples of using the above methods to identify perturbations as those that affect specific cell state transitions are provided below in Section IV.E.

[0230] III.E. Methods for Identifying the Biological Utility of Perturbations

[0231] In some embodiments, the disclosed methods are used to identify the biological effects of a perturbation. These methods encompass measurements of any cellular component (or combination of different cellular components) that can be shown to be differentially present in cells with different states or phenotypes (e.g., diseased and normal phenotypes). That is, the presence, absence, or amount of a cellular component is associated with a cell state or phenotype. In one embodiment, the method comprises: exposing a plurality of cells to a perturbation; performing a first differential cellular component expression assay, the assay comprising accessing a first plurality of single cell expression data sets obtained from the cells before and after the plurality of cells are exposed to the perturbation, each data set comprising a vector r of a cellular component. i, each entry in the vector is associated with one of a plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell; applying a statistical technique to the first plurality of data sets to generate differentially expressed cellular components E in response to exposure to the perturbation k and determining the differentially expressed cellular components E in response to exposure to the perturbation k The set and differentially expressed cellular components associated with the differences between diseased and normal cell phenotypes l The similarity level between sets of k and E l The significant similarity level between the diseased cell phenotype and the normal cell phenotype indicates the utility of the perturbation in the transition of the cell between the diseased cell phenotype and the normal cell phenotype.

[0232] In some embodiments, applying the statistical technique includes performing dimensionality reduction on the first plurality of data sets 132 (e.g., Figure 3 Dimensionality reduction 320) to generate a first matrix M, the first matrix M includes rows in the first dimension and columns in the second dimension, and the values ​​of the matrix M include values ​​generated by the amount of the cell component at the point located in the first dimensional space and the second dimensional space; clustering is performed to generate clusters C j A first set of clusters C, each cluster including a plurality of points corresponding to a subset of rows in the first matrix M and their corresponding cell response states; and using the clusters C j The first set identifies the cellular components that are differentially expressed by the cells in response to exposure to the perturbation E k of collection.

[0233] In some embodiments, the differentially expressed cellular component E associated with the difference between the diseased cell phenotype and the normal cell phenotype l The set of differential cellular component expression data may be determined by performing a second differential cellular component expression assay, the second assay comprising accessing a second plurality of single cell cellular component expression data sets obtained from a plurality of cells in different states, such as normal cells and diseased cells, each data set comprising a vector r of cellular components. i , each entry in the vector is associated with one of a plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell; and applying a statistical technique to the second plurality of data sets.

[0234] In some embodiments, applying the statistical technique to the second plurality of data sets comprises: performing dimensionality reduction on the second plurality of data sets to generate a second matrix M, the second matrix M comprising rows in a first dimension and columns in a second dimension, the values ​​of the second matrix M comprising values ​​generated from the amounts of one or more cellular components at points located in the first dimensional space and the second dimensional space; performing manifold learning under a relative similarity approximation of the points with the second matrix M to create a second matrix N comprising a plurality of rows and two columns, each row corresponding to one of the cells, each column corresponding to one of the two dimensions in the two-dimensional space, the values ​​of the second matrix N indicating the relative difference in the cellular phenotype between each cell relative to each other cell based on the data set; performing clustering to generate clusters C j a second set of clusters C, each cluster including a plurality of points corresponding to a subset of the rows in the matrix N and their corresponding cell response states; and using clusters C j The second set identifies differentially expressed cellular components associated with differences between the diseased cell phenotype and the normal cell phenotype in the cell. l A set of markers that indicate differences between the diseased cell phenotype and the normal cell phenotype.

[0235] In some embodiments, the perturbation is known to have an acceptable human safety profile as determined by results obtained in regulated clinical trials.

[0236] In some embodiments, the diseased cell phenotype is identified by the difference between the diseased cell and the normal cell. For example, in some embodiments, it can be identified by the following: loss of cell function, acquisition of cell function, progression of the cell (e.g., the transition of the cell to a differentiated state), stagnation of the cell (e.g., the cell cannot transition to a differentiated state), invasion of the cell (e.g., the cell appears in an abnormal position), disappearance of the cell (e.g., the cell does not exist at the position where the cell normally exists), disorder of the cell (e.g., structural, morphological and / or spatial changes in and / or around the cell), loss of the cell network (e.g., cell changes that eliminate normal effects in daughter cells or downstream cells of the cell), acquisition of the cell network (e.g., cell changes that trigger new downstream effects in the daughter cells of the downstream cells of the cell), excess of the cell (e.g., the cell is too abundant), deficiency of the cell (e.g., the cell density is below a critical threshold), differences in the ratio and / or amount of cell components in the cell, differences in the rate of transition in the cell, or any combination thereof.

[0237] In some embodiments, diseased cells include cell lines, biopsy cells, and cultured primary cells. In some embodiments, normal cells include cultured primary cells and biopsy cells. In some embodiments, cells are human cells.

[0238] In some embodiments, the method is used to select a perturbation that can be used to treat a disease based on the indicated utility identified using the above method. In some embodiments, the method comprises treating a subject suffering from a disease by administering to the subject an effective amount of a selected perturbation or a drug substance developed from a perturbation lead compound.

[0239] Implementation

[0240] Embodiment 1. A method comprising the steps of: accessing a plurality of single cell cell component expression datasets, each dataset being obtained from one of a plurality of cells that have transformed from the same "progenitor" cell type, each dataset comprising a vector r of a cell component i , each entry in the vector is associated with one of the plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell; performing dimensionality reduction on the data set to generate a matrix M, the matrix M comprising rows in a first dimension and columns in a second dimension, each row corresponding to one of the plurality of cells, the values ​​of the matrix M comprising values ​​generated by the amount of the cellular component at a point located in the first dimensional space and the second dimensional space; performing clustering to generate clusters C j A set of clusters C, each cluster including a plurality of points corresponding to a subset of the rows in the matrix M and their corresponding cells; and using the clusters C j The set of cells that differentially express cellular components E k of collection.

[0241] Embodiment 2. A method as described in Embodiment 1, which also includes performing manifold learning under the relative similarity approximation of points with the matrix M to create a matrix N including multiple rows and two columns, each row corresponding to one of the multiple cells, and each column corresponding to one of the two dimensions in the two-dimensional space, and the values ​​of the matrix N indicate the relative cell type of each cell relative to each other cell based on the data set.

[0242] Embodiment 3. A method as described in any one of Embodiments 1 to 2, wherein when the single-cell cell component expression dataset is obtained, the cells are a heterogeneous cell population having various cell types.

[0243] Embodiment 4. A method as described in any of Embodiments 1 to 2, wherein the cells are a substantially homogeneous cell population having the "progenitor" cell type; and wherein the single-cell cellular component expression dataset is obtained at each of multiple time points when the cells transition from the "progenitor" cell type, so that a different dataset among the multiple datasets is collected for each cell and time point combination.

[0244] Embodiment 5. A method as described in Embodiment 4, wherein the multiple time points include at least three time points.

[0245] Embodiment 6. A method as described in any of Embodiments 4 to 5, wherein the multiple time points include "progenitor" time points at which a substantial portion of the cells have not yet transformed from the "progenitor" cell type.

[0246] Embodiment 7. A method as described in any one of Embodiments 4 to 6, wherein the multiple time points include transition time points at which a substantial portion of cells have transitioned from the "progenitor" cell type.

[0247] Embodiment 8. A method as described in any one of Embodiments 4 to 7, wherein the multiple time points include at least one intermediate time point at which a substantial portion of cells have at least partially transformed from the "progenitor" cell type.

[0248] Embodiment 9. The method of any one of embodiments 1 to 8, wherein the plurality of cellular components is selected from the group consisting of nucleic acids, proteins, lipids, carbohydrates, nucleotides, and any combination thereof.

[0249] Embodiment 10. The method of embodiment 9, wherein the nucleic acid is selected from the group consisting of DNA and RNA.

[0250] Embodiment 11. A method as described in Embodiment 10, wherein the RNA is selected from the group consisting of coding RNA and non-coding RNA.

[0251] Embodiment 12. A method as described in any of embodiments 1 to 11, wherein the single-cell cellular component expression dataset is generated using a method selected from the group consisting of: single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq and any combination thereof, and summaries thereof, including combinations representing activated pathways in the single-cell cellular component expression dataset, such as linear combinations.

[0252] Embodiment 13. A method as described in any one of Embodiments 1 to 12, wherein performing dimensionality reduction includes performing principal component analysis (PCA) on the single-cell cellular component expression dataset to generate the matrix M.

[0253] Embodiment 14. A method as described in any one of Embodiments 1 to 13, wherein performing dimensionality reduction includes using diffusion mapping on the single-cell cellular component expression dataset to generate the matrix M.

[0254] Embodiment 15. A method as described in any one of Embodiments 1 to 14, wherein performing dimensionality reduction includes using a neural network autoencoder on the single-cell cellular component expression dataset to generate the matrix M.

[0255] Embodiment 16. A method as described in Embodiment 2, wherein performing manifold learning includes estimating the geometry of the data in the matrix M to create the matrix N.

[0256] Embodiment 17. A method as described in Embodiment 16, wherein performing manifold learning includes performing local linear embedding.

[0257] Embodiment 18. A method as described in Embodiment 16, wherein performing manifold learning includes performing local linear isometric mapping (ISOMAP).

[0258] Embodiment 19. A method as described in Embodiment 16, wherein performing manifold learning includes performing t-distributed stochastic neighbor embedding (t-SNE).

[0259] Embodiment 20. A method as described in Embodiment 16, wherein performing manifold learning includes performing affinity-based thermal diffusion potential for trajectory embedding (PHATE).

[0260] Embodiment 21. A method as described in Embodiment 16, wherein performing manifold learning includes performing unified manifold approximation and projection (UMAP).

[0261] Embodiment 22. A method as described in Embodiment 16, wherein performing manifold learning includes creating a force-directed layout.

[0262] Embodiment 23. A method as described in Embodiment 22, wherein the force-directed layout is created using the Force Atlas 2 algorithm.

[0263] Embodiment 24. A method as described in any of Embodiments 1 to 23, wherein clustering is performed assuming no prior knowledge about the organization of the multiple points in each cluster.

[0264] Embodiment 25. A method as described in any one of embodiments 1 to 24, wherein performing clustering includes performing HDBSCAN to generate the cluster C j of collection.

[0265] Embodiment 26. A method as described in any one of embodiments 1 to 25, wherein performing clustering includes performing Louvain community detection to generate clusters C j of collection.

[0266] Embodiment 27. The method of any one of embodiments 1 to 26, wherein performing clustering comprises assigning each point to a cluster C based on the time point at which the single-cell cellular component expression dataset associated with each point was collected. j one of the.

[0267] Embodiment 28. A method as described in any of embodiments 1 to 27, wherein performing clustering includes using a diffusion path algorithm to analyze the multiple points, and the diffusion path algorithm assigns the points to the clusters based on a measure of the degree to which the points are cluster ends.

[0268] Embodiment 29. The method of any one of embodiments 1 to 28, wherein the differentially expressed cellular component E is determined k The set comprises: for each cellular component, for at least one of the clusters, comparing the amount of the cellular component at the plurality of points in the at least one cluster with the amount of the cellular component at the plurality of points in at least one other cluster; and in response to the amount of the cellular component at the plurality of points in the at least one cluster being greater than a threshold level of the amount of the cellular component at the plurality of points in the at least one other cluster, adding the cellular component to the differentially expressed cellular component E k of concentration.

[0269] Embodiment 30. A method as described in embodiment 29, wherein the at least one cluster includes the cluster C j A lineage cluster containing multiple spots that can be identified as having the desired cell type.

[0270] Embodiment 31. A method as described in embodiment 30, wherein the at least one other cluster includes the cluster C j Out-of-lineage clusters in the , which contain spots that can be identified as having undesirable cell types.

[0271] Embodiment 32. The method of any one of embodiments 1 to 31, wherein the differentially expressed cellular component E is determined k The set comprises: for each cellular component, for at least one of the clusters, calculating a distance metric between the amount of the cellular component at the plurality of points in the at least one cluster and the amount of the cellular component at the plurality of points in the at least one other cluster; and in response to the distance metric being statistically significant, adding the cellular component to the differentially expressed cellular components E k of concentration.

[0272] Embodiment 33. The method of any one of embodiments 1 to 32, further comprising screening the differentially expressed cellular component E against a transcription factor database.k to identify sets of differentially expressed transcription factors.

[0273] Embodiment 34. A method as described in embodiment 33, further comprising: k The method further comprises performing empirical mode decomposition on the set of differentially expressed transcription factors to generate a pseudo-time representation of the data set; and identifying the set of differentially expressed transcription factors based on the pseudo-time representation.

[0274] Embodiment 35. A method comprising the steps of: accessing a plurality of single cell cell component expression datasets, each dataset being obtained from one of a plurality of cells that have transformed from the same "progenitor" cell type, each dataset comprising a vector r of a cell component i , each entry in the vector is associated with one of the plurality of cellular components, and the value of each entry represents the amount of the cellular component of the cell; using a kNN algorithm and using the single cell cellular component expression data set to generate a kNN graph; performing clustering to generate a cluster C j A set of clusters, each cluster comprising a plurality of points, each point corresponding to a single cell cell component expression dataset of one cell in the plurality of cells; and using the cluster C j The set of cells that differentially express the cellular components E k of collection.

[0275] Embodiment 36. A method as described in embodiment 35, wherein the differentially expressed cellular component E is determined k The set includes the clusters C j The distance measure between the multiple points in .

[0276] Embodiment 37. A method comprising the following steps: accessing a single-cell transition feature, wherein the single-cell transition feature represents a measure of the differential cellular component expression between a first cell state and an altered cell state; accessing a perturbation feature, wherein the perturbation feature represents a measure of the differential cellular component expression between an unperturbed cell that is not exposed to the perturbation and a perturbed cell that is exposed to the perturbation; and based on a comparison of the single-cell transition feature with the perturbation feature, determining whether the perturbation is associated with a transition of the cell between the first cell state and the altered cell state.

[0277] Embodiment 38. A method as described in Embodiment 36, wherein accessing the single-cell transition feature includes: determining the single-cell transition feature based on a first plurality of single-cell cell component expression data sets and providing a second plurality of single-cell cell component expression data sets, each first data set is obtained from a cell in a first plurality of cells in the first cell state, and each second data set is obtained from a cell in a second plurality of cells in the changed cell state.

[0278] Embodiment 39. A method as described in embodiment 38, wherein each of the first plurality of single-cell cellular component expression datasets and the second plurality of single-cell cellular component expression datasets comprises a vector r of cellular components. i , each entry in the vector is associated with one of a plurality of cellular components, and the value of each entry represents the amount of the cellular component in the cell.

[0279] Embodiment 40. A method as described in any one of embodiments 38 to 39, which also includes: obtaining the first plurality of single-cell cell component expression datasets and the second plurality of single-cell cell component expression datasets, wherein the obtaining includes: performing dimensionality reduction on the first plurality of single-cell cell component expression datasets and the second plurality of single-cell cell component expression datasets to generate a matrix M, wherein the matrix M includes rows in a first dimension and columns in a second dimension, each row corresponds to a cell in the plurality of cells, and the values ​​of the matrix M include values ​​generated by the amounts of cell components at points located in the first dimensional space and the second dimensional space; performing clustering to generate clusters C j A set of each cluster including a plurality of points corresponding to a subset of the rows in the matrix M and their corresponding cells; from the cluster C j The first cluster of the set identifies the first plurality of cells; from the cluster C j identifying the second plurality of cells by a second cluster of the set; obtaining the first plurality of single-cell cellular component expression datasets from the first plurality of cells; and obtaining the second plurality of single-cell cellular component expression datasets from the second plurality of cells.

[0280] Embodiment 41. A method as described in Embodiment 40, which also includes performing manifold learning under relative similarity approximation of points with the matrix M to create a matrix N including multiple rows and two columns, each row corresponding to a cell in the first multiple cells and the second multiple cells, and each column corresponding to one of the two dimensions in the two-dimensional space, and the values ​​of the matrix N indicate the relative cell state of each cell relative to each other cell based on the first multiple single-cell cell component expression datasets and the second multiple single-cell cell component expression datasets.

[0281] Embodiment 42. A method as described in any one of embodiments 40 to 41, wherein the steps are performed according to any one of the methods of embodiments 1 to 34.

[0282] Embodiment 43. A method as described in any one of Embodiments 37 to 42, wherein accessing the perturbation characteristic comprises: determining the perturbation characteristic based on multiple unperturbed single-cell cellular component expression datasets of the unperturbed cells that are not exposed to the perturbation and based on multiple perturbed single-cell cellular component expression datasets of the perturbed cells that are exposed to the perturbation.

[0283] Embodiment 44. A method as described in any of Embodiments 37 to 43, wherein the unperturbed cells are control cells that have not been exposed to the perturbation of the perturbed cells, or wherein the unperturbed cells are the average of unrelated perturbed cells that have been exposed to the perturbation.

[0284] Embodiment 45. The method of any one of Embodiments 37 to 44, further comprising the step of filtering the single-cell transition signature and the perturbation signature to include cellular components that are transcription factors.

[0285] Embodiment 46. A method as described in any one of embodiments 38 to 42, wherein determining the single-cell transition features based on the first plurality of single-cell cell component expression datasets and the second plurality of single-cell cell component expression datasets includes: using one of a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model to determine the difference in cellular component amounts between the first plurality of single-cell cell component expression datasets and the second plurality of single-cell cell component expression datasets.

[0286] Embodiment 47. A method as described in Embodiment 43, wherein determining the disturbance feature based on the unperturbed multiple single-cell cellular component expression datasets and the perturbed multiple single-cell cellular component expression datasets includes: using one of a mean difference test, a Wilcoxon rank sum test (Mann-Whitney U test), a t-test, a logistic regression, and a generalized linear model to determine the difference in cellular component amounts between the unperturbed multiple single-cell cellular component expression datasets and the perturbed multiple single-cell cellular component expression datasets.

[0287] Embodiment 48. The method as described in any one of Embodiments 37 to 47, further comprising: filtering the single-cell transition features and the perturbation features to reduce the number of cellular components included in the single-cell transition features and the perturbation features.

[0288] Embodiment 49. A method as described in Embodiment 48, wherein filtering the single-cell transition features and the perturbation features includes reducing the number of cellular components included in the single-cell transition features and the perturbation features according to a threshold p value or according to a threshold number of cellular components.

[0289] Embodiment 50. A method as described in any of embodiments 37 to 49, wherein the perturbation feature includes multiple cellular components, each cellular component is associated with a significance score, the significance score quantifies the association between the change in the amount of the cellular component and the change in the cellular state between the unperturbed cell and the perturbed cell, and wherein determining whether the perturbation is associated with the transition of the cell between the first cellular state and the changed cellular state includes: replacing the significance score of each cellular component with the matching score of the cellular component; combining the matching scores of the multiple cellular components to generate a matching score for the perturbation; and determining whether the perturbation is associated with the transition of the cell between the first cellular state and the changed cellular state based on the matching score of the perturbation.

[0290] Embodiment 51. A method as described in embodiment 50, wherein the matching score comprises a discrete score or a continuous score.

[0291] Embodiment 52. A method as described in any one of embodiments 50 to 51, wherein replacing each significant score comprises: if the amount of the cellular component of the cellular component from the single-cell transition feature and the amount of the cellular component from the perturbation feature are both upregulated, replacing the significant score with a first score; if the amount of the cellular component of the cellular component from the single-cell transition feature is upregulated and the amount of the cellular component from the perturbation feature is downregulated, replacing the significant score with a second score; and if the amount of the cellular component of the cellular component from the perturbation feature is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0292] Embodiment 53. A method as described in any one of Embodiments 50 to 51, wherein replacing the significant score comprises: if the amount of the cellular component from the single-cell transition feature and the amount of the cellular component from the perturbation feature of the cellular component are both downregulated, replacing the significant score with a first score; if the amount of the cellular component from the single-cell transition feature of the cellular component is downregulated and the amount of the cellular component from the perturbation feature is upregulated, replacing the significant score with a second score; and if the amount of the cellular component from the perturbation feature of the cellular component is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0293] Embodiment 54. A method as described in any of Embodiments 37 to 49, wherein the perturbation feature includes multiple cellular components, each cellular component is associated with a significance score, the significance score quantifies the association between the change in the amount of the cellular component and the change in the cellular state between the unperturbed cell and the perturbed cell, and wherein determining whether the perturbation is associated with the transition of the cell between the first cellular state and the changed cellular state includes: combining the significance scores of the multiple cellular components to generate a significance score for the perturbation; and based on the significance score of the perturbation, determining whether the perturbation is associated with the transition of the cell between the first cellular state and the changed cellular state.

[0294] Embodiment 55. A method as described in any one of embodiments 50 to 53, which also includes: estimating the erroneous cellular component discovery rate of the matching score of the perturbation by: calculating the empirical marginal expression frequency of each of the multiple cellular components; summing the empirical marginal expression frequencies of the multiple cellular components over their combination to generate a probability of identifying many cellular components by chance assuming independent distribution of expression; and estimating the erroneous cellular component discovery rate of the matching score of the perturbation based on the probability.

[0295] Embodiment 56. A method as described in any of Embodiments 37 to 55, wherein determining whether the perturbation is associated with the transition of the cell between the first cell state and the changed cell state comprises: determining that a threshold amount of a covariate of the perturbation is associated with the transition of the cell between the first cell state and the changed cell state; and in response to the determination, determining that the perturbation is associated with the transition of the cell between the first cell state and the changed cell state.

[0296] Embodiment 57. A method as described in Embodiment 56, wherein the perturbation comprises exposing the cell to a small molecule, and wherein the one or more covariates of the perturbation comprise: a specific dose of the small molecule, the time at which the differential expression of cellular components between the unperturbed cell and the perturbed cell is measured relative to the time at which the perturbed cell is exposed to the small molecule, and the cell line of the perturbed cell.

[0297] Embodiment 58. The method of any one of Embodiments 37 to 57, wherein the cellular components comprise genes.

[0298] Embodiment 59. A method as described in embodiments 37 to 58, wherein the single-cell cellular component expression dataset is generated using a method selected from the group consisting of: single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination or summary thereof.

[0299] Embodiment 60. A method as described in embodiments 37 to 59, wherein at least one of the single-cell transition signature and the perturbation signature is obtained from a database.

[0300] Embodiment 61. A method as described in Embodiment 60, wherein the perturbation signature is obtained from a database of multiple perturbation signatures of multiple perturbations, and wherein the method further comprises: for each perturbation of the multiple perturbations in the database: accessing the perturbation signature of the perturbation from the database; and determining whether the perturbation is associated with the transition of the cell between the first cell state and the changed cell state based on a comparison of the single-cell transition signature with the perturbation signature.

[0301] Embodiment 62. A method as described in any one of Embodiments 37 to 61, which also includes accessing multiple perturbation characteristics of multiple disturbed cells; and for each of the multiple perturbation characteristics, performing the determination step to screen for perturbations that promote the changed cell state.

[0302] Embodiment 63. A method as described in Embodiment 62, wherein accessing the multiple perturbation characteristics includes exposing cells to multiple perturbations to generate the multiple perturbed cells; and measuring multiple cellular component quantities from the multiple perturbed cells.

[0303] Embodiment 64. The method of any one of Embodiments 37 to 63, further comprising identifying a perturbation that promotes the altered cellular state.

[0304] Embodiment 65. A method as described in Embodiment 64, wherein promoting the altered cell state includes, in a cell population including the first cell state, promoting a transition from the first cell state to the altered cell state.

[0305] Embodiment 66. A method as described in Embodiment 64, wherein promoting the altered cell state includes, in a cell population including the first cell state, increasing the ratio of the number of cells in the alternative state to the number of cells in the first state or a state optionally different from the altered cell state.

[0306] Embodiment 67. A method as described in Embodiment 64, wherein promoting the altered cell state includes increasing the absolute number of cells in the altered cell state in a cell population including the first cell state.

[0307] Embodiment 68. A method as described in Embodiment 64, wherein promoting the altered cell state includes, in a cell population including the first cell state, reducing the absolute number of cells in the first cell state or a state optionally different from the altered cell state.

[0308] Embodiment 69. A method as described in any one of Embodiments 37 to 68, wherein the cell transformation signature and the perturbation signature are generated using different types of cellular components.

[0309] Embodiment 70. A method as described in any one of Embodiments 37 to 68, wherein the cell transformation signature and the perturbation signature are generated using the same type of cellular components.

[0310] Embodiment 71. A method comprising the following steps: accessing a single-cell transition signature, wherein the single-cell transition signature represents a measure of differential cellular component expression between a first cell state and an altered cell state; accessing a plurality of perturbation signatures, each perturbation signature being associated with a perturbation and representing a measure of differential cellular component expression between an unperturbed cell that is not exposed to the perturbation and a perturbed cell that is exposed to the perturbation; and determining a subset of the perturbations associated with the transition of a cell between the first cell state and the altered cell state based on a comparison of the single-cell transition signature with the plurality of perturbation signatures.

[0311] Embodiment 72. A method as described in embodiment 71, wherein each perturbation feature includes multiple cellular components, each cellular component is associated with a significance score, the significance score quantifies the association between the change in the amount of the cellular component and the change in the cellular state between the unperturbed cell and the perturbed cell, and wherein determining the subset of the perturbations associated with the transition of the cell between the first cellular state and the changed cellular state includes: for each perturbation feature: replacing the significance score of the cellular component with the matching score of each cellular component; and combining the matching scores of the multiple cellular components to generate a matching score for the perturbation; sorting the perturbations based on the matching scores of the perturbations; and selecting the subset of the perturbations based on the sorted list of the perturbations.

[0312] Embodiment 73. A computer program product comprising a non-transitory computer-readable storage medium having instructions encoded thereon, which instructions, when executed by a processor, cause the processor to perform the method as described in any one of Embodiments 37 to 72.

[0313] Embodiment 74. A system comprising: a non-transitory computer-readable storage medium having instructions encoded thereon, the instructions, when executed by a processor, causing the processor to perform the method as described in any one of Embodiments 37 to 72.

[0314] Embodiment 75. A method for promoting neurons and / or progenitor cells, comprising: exposing a starting cell population including fibroblasts to a perturbation having a perturbation characteristic, wherein the perturbation characteristic promotes the transformation of the starting cell population including fibroblasts into progenitor cells and / or neurons, wherein the perturbation characteristic is an increase in the activity of one or more of Brn2, Ascl1, Myt1, Zfp941, Taf5B, St18, Zkscan16, Camta1 and Arnt2 and / or a decrease in the activity of one or more of Ascl1, Atf3, Rorc, Scx, Satb1, Elf3 and Fos.

[0315] Embodiment 76. A method as described in Embodiment 75, wherein the neurons and / or progenitor cells are promoted by one or more of the following: increasing the absolute number of neurons and / or progenitor cells; reducing the absolute number of fibroblasts; promoting the transformation of fibroblasts into neurons and / or progenitor cells; promoting the lifespan of neurons or progenitor cells; reducing the lifespan of fibroblasts; or increasing the ratio of neurons and / or progenitor cells to fibroblasts.

[0316] Embodiment 77. A method as described in Embodiment 75, wherein the perturbation does not include forskolin, PP1, PP2 and trichostatin A.

[0317] Embodiment 78. A method for increasing the amount of neurons and / or progenitor cells, comprising exposing a cell population comprising fibroblasts to a pharmaceutical composition having a perturbation characteristic that promotes the conversion of the cell population comprising fibroblasts into neurons, wherein the pharmaceutical composition comprises forskolin, PP1, PP2, trichostatin A, BRD-K38615104, geldanamycin, manumitin A, mitoxantrone, curcumin, avosidide, varinostat, KI20227, or a combination of the foregoing (e.g., 2, 3, 4, 5 or more of the foregoing).

[0318] Embodiment 79. The method of embodiment 78, wherein the pharmaceutical composition does not contain forskolin, PP1, PP2, and trichostatin A.

[0319] Embodiment 80. A pharmaceutical composition for promoting neurons and / or progenitor cells, comprising: a perturbation selected from the group consisting of forskolin, PP1, PP2, trichostatin A, BRD-K38615104, geldanamycin, manumitin A, mitoxantrone, curcumin, avosidide, varinostat, KI20227 or a combination of the foregoing substances; and a pharmaceutically acceptable excipient.

[0320] Embodiment 81. The pharmaceutical composition of embodiment 80, wherein the perturbation does not include forskolin, PP1, PP2, and trichostatin A.

[0321] Embodiment 82. A unit dosage form comprising the pharmaceutical composition of embodiment 80 or 81.

[0322] Embodiment 83. A method for identifying a candidate perturbation for promoting the transition of a starting cell population including fibroblasts to neurons and / or progenitor cells, the method comprising: exposing the starting cell population including fibroblasts to a perturbation; identifying a perturbation feature of the perturbation, the perturbation feature comprising one or more cellular components and a significance score associated with each cellular component, the significance score of each cellular component quantifying the association between a change in expression of the cellular component after the cell population is exposed to the perturbation and a change in the cell state of the cell population from fibroblasts to neurons and / or progenitor cells; and identifying the perturbation as a candidate perturbation for promoting the transition of a cell population including fibroblasts to neurons and / or progenitor cells based on the perturbation feature, wherein the perturbation feature is an increase in the activity of one or more of Brn2, Ascl1, Myt1, Zfp941, Taf5B, St18, Zkscan16, Camta1 and Arnt2 and / or a decrease in the activity of one or more of Ascl1, Atf3, Rorc, Scx, Satb1, Elf3 and Fos.

[0323] IV. Examples 0, 1, 2 and 3 - Identification of Factors in Mouse Embryonic Fibroblasts Differentiating into Neurons and Myocytes Effects and control of cell fate

[0324] The following examples verify the methods described in Part II and Part III above. In more detail, the examples show that the methods of Part II and Part III accurately identify genes and / or perturbations known to affect cell state transition trajectories. In addition, the examples discussed below show that the methods of Part II and Part III generate the ability to control new biological understandings of cell state transition trajectories. Specifically, the examples show that the methods of Part II and Part III identify previously unknown factors (e.g., genes and perturbations) that affect cell state transitions.

[0325] The examples discussed below apply the methods of Sections II and III to a combination of publicly available data and in vitro experimental data to validate several known and previously unknown factors (e.g., genes and perturbations) that affect cell state transition trajectories. The results of applying the methods of Sections II and III to a combination of publicly available data and in vitro experimental data are shown in FIG. 4B to FIG. 5A and FIG. 7A to FIG. 9 middle.

[0326] Some of these results were also validated using only in vitro experimental data. The results of this in vitro validation are shown in Figure 6 In vitro experimental data were obtained by growing and measuring cells according to the protocol discussed in Section IV.A below.

[0327] IV.A. Example 0: In vitro cell processing and data set acquisition

[0328] This section describes the protocol used for the in vitro experiments mentioned above. Data from this in vitro experiment were combined with publicly available data to generate FIG. 4B to FIG. 5A and FIG. 7A to FIG. 9 , and independently used to generate Figure 6 .

[0329] This section applies the general protocol described in Section II to a specific embodiment of evaluating mouse embryonic fibroblasts (MEFs) that differentiate into neurons or muscle cells. In this specific embodiment, neurons are lineage-internal cells, muscle cells are lineage-external cells, and MEFs are "progenitor" cells. The protocol also includes additional steps, including lentiviral overexpression of the gene Ascl1 and perturbation mediation.

[0330] MEF medium is 10% fetal bovine serum (FBS), 1x Glutamax, 1x non-essential amino acids, Pen / Strep and β-mercaptoethanol in Dulbecco's Modified Eagle Medium (DMEM). Neuronal medium is DMEM / F12, N2, B27, 1x Glutamax and 25 μg / ml insulin.

[0331] The protocols followed are listed below:

[0332] Day 0: Thaw 1 million MEF cells into a 10 cm plate in MEF medium.

[0333] Day 1: Seed into 24-well plates at 20K / well.

[0334] If applicable, the cells were simultaneously spin infected with Ascl1 virus (MOI 8) in the presence of MEF medium (250 μl / well) and 8 μg / ml polybrene at 2000 rpm for 1 hour at 32°C.

[0335] Single-cell RNA (RNA) sequencing (scRNA-seq) was performed to obtain a d2 dataset for each cell.

[0336] Day 2: Change the medium to wash off the polybrene used for virus experiments (MEF medium).

[0337] For perturbation experiments, small molecules (resuspended in dimethyl sulfoxide (DMSO) or ethanol) were added.

[0338] Day 3: Change the culture medium to neuronal culture medium.

[0339] For perturbation experiments, molecules were added (resuspended in DMSO or ethanol).

[0340] Day 5: Half medium change (addition of small molecules if applicable)

[0341] Day 8: Half medium change (addition of small molecules if applicable)

[0342] Day 9: Half medium change (addition of small molecules if applicable)

[0343] Day 11: Half medium change (addition of small molecules if applicable)

[0344] Day 13: Half medium change (addition of small molecules if applicable)

[0345] Day 15: Fix the plates and stain with Map2 and Tuj1 antibodies. Image on a Molecular Devices HCI IXM4 or other high-content imaging microscope by scanning each well. Quantify the number of Map2 / Tuj1 positive neurons per well.

[0346] IV.B. Examples 1, 2, and 3: In vitro cell processing and data set acquisition

[0347] Figure 4A A timeline depicting the trajectory of an induced cell state transition over a period of time according to one embodiment. More specifically, Figure 4A Depicted is a timeline tracking the trajectory of induced MEF transformation over a 23 day period (day 0 to day 22).

[0348] like Figure 4AAs shown in , at day 0 of a 23-day period, MEFs were obtained. In alternative embodiments, the transformation trajectory of any single cell can be studied according to similar methods. For example, in alternative embodiments, the transformation trajectory of mouse embryonic hematopoietic cells can be studied according to similar methods.

[0349] On day 0 of the 23-day period, each MEF in the MEF population is transduced with one or more appropriate transcription factors. Figure 4A As shown in FIG, only Ascl1 or Brn2, Ascl1 and Myt11 (collectively referred to as BAM transcription factors) were overexpressed in MEFs. Specifically, in the generation of Figure 6 In the in vitro experiments, only Ascl1 was overexpressed in MEFs. In contrast, to generate FIG. 4B to FIG. 5A and FIG. 7A to FIG. 9 Based on publicly available data, each of Brn2, Ascl1, and Myt11 was overexpressed in MEFs.

[0350] In embodiments disclosed herein, following lentiviral delivery, inducible expression of Ascl1 forces expression of the Ascl1 transcription factor. In alternative embodiments, any alternative means can force expression of one or more transcription factors. For example, in alternative embodiments, transposons, mRNA delivery, or another type of viral delivery can force expression of one or more transcription factors.

[0351] It is known that the forced expression of one or more of the BAM transcription factors causes one or more of the forced MEFs to be more generally transformed into mouse "progenitor" cells, mouse neurons and / or mouse muscle cells. Specifically, as known in the literature, Ascl1 triggers the induction of MEFs to be transformed into mouse "progenitor" cells, the expression of Ascl1 alone induces mouse "progenitor" cells to be transformed into mouse neurons and mouse muscle cells, and the expression of Brn2 and Myt1l induces mouse "progenitor" cells to be transformed into mouse neurons. However, this induction of cell state transformation by one or more of the BAM transcription factors does not occur with 100% efficiency. Specifically, as known in the literature, the BAM transcription factors induce the transformation of MEF to mouse neurons with an efficiency of 20%. In other words, despite expressing one or more of the BAM transcription factors, some cells may not be transformed as expected. In some embodiments, this failed transformation is referred to as failed reprogramming.

[0352] Mouse cells in which one or more of the BAMs are forced to be expressed are monitored over a 23 day period. More specifically, for mouse cells in which Ascl1 expression is forced, single cell RNA sequencing (scRNA-seq) measurements are obtained for each single mouse cell of the mouse cells in the population on days 2, 5, and 22 during the 23 day period. Alternatively, for mouse cells in which expression of all BAM factors is forced, scRNA-seq measurements are obtained for each single mouse cell of the mouse cells in the population only on day 22 during the 23 day period.

[0353] In an alternative embodiment, RNA sequencing measurements can be performed at any frequency at any number of time points. More specifically, in order to accurately capture the cell state transition trajectory, the time point for RNA sequencing measurements is ideally usually corresponding to the time point at which one or more transition trajectories diverge. RNA sequencing measurements are performed on single cells on a specific day, including mRNA expression in the quantified single cell on the specific day. In other words, RNA sequencing measurements are performed on single cells on a specific day, including counting of each mRNA transcript in the single cell on the specific day. In addition, because each mRNA transcript is associated with a specific gene, the RNA sequencing measurements of single cells on a specific day are included in the quantification of gene expression in the single cell on the specific day. However, in practice, cells will generally not be completely homogeneous in the state of their cell state transition, and therefore it is predicted that the measurement of cell state transition on a given day is to capture the distribution of cells in each stage of cell state transition.

[0354] The in vitro protocol for overexpressing Ascl1 in MEFs was used to perform Figure 6 In addition, gene expression measurements obtained from the in vitro protocol overexpressing Ascl1 in MEFs were combined with publicly available gene expression measurements from MEFs overexpressing all BAM factors. The in vitro and publicly available data pools were then used to generate FIG. 4B to FIG. 5A and FIG. 7A to FIG. 9 As described above, these figures serve both to validate the ability of the methods of Parts II and III to accurately identify genes that affect cell state transitions and to demonstrate the ability of the methods described in Parts II and III to generate new biological understanding that can be used to control the trajectories of cell state transitions and, therefore, cell fate.

[0355] IV.C. Example 1: Dimensionality reduction of transformations identified by validation literature

[0356] As discussed above, gene expression measurements obtained at days 2, 5, and 22 from MEFs overexpressing only Ascl1 were combined with publicly available gene expression measurements obtained from MEFs overexpressing all BAM factors at day 22. Using the methods described above in Part II, for each day that gene expression in cells was measured, the gene expression measurements for each cell were used to generate a transcript vector r i Each transcript vector r i With the transcript vector r i The transcript vector r contains genes whose expression was measured on a specific day in a specific cell. i Each transcript in is associated with a specific gene in the cell genome, and the transcript vector r i Each entry in represents the value of the transcript vector r i The sequencing depth (transcript count) of the associated transcripts on a specific day.

[0357] As discussed above with respect to Section III.C., dimensionality reduction is performed on a dataset encoding gene expression measurements for each cell on each measurement day. In this embodiment, principal component analysis (PCA) is used to perform dimensionality reduction and generate a reduced matrix M.

[0358] Next, manifold learning is performed on the matrix M to generate a further reduced-dimensional matrix N. In this embodiment, a force-directed layout algorithm is used to generate the matrix N. The matrix N is depicted in Supplementary Table 1. The matrix N is also plotted as Figure 4B The force-directed layout manifold depicted in . Figure 4B The plotted data in the manifold of corresponds to the matrix N data in Supplementary Table 1. Note that the matrix N is mainly used for visualization purposes and does not need to be generated in some embodiments. In other words, in some embodiments, manifold learning is not performed on the matrix M.

[0359] As discussed above, each point in the manifold is associated with a row of the matrix N, which is associated with a specific cell in the cells on a specific day of the four days on which the cell's gene expression was measured. In addition, each point is associated with a dataset of gene transcript counts measured for a specific cell on a specific day. Figure 4B When the manifold is shown, because the values ​​of dimension x and dimension y in a row of the manifold are based on the gene transcript counts of the cells on that day associated with the row, the positioning of a point in the manifold relative to other points, and therefore relative to other cells on other days, reflects the gene transcript counts of the cells on that day associated with the point in the manifold. Thus, visual inspection of the manifold allows observation of the changing gene transcript counts of various genes of the cells over a 23-day period.

[0360] exist Figure 4BIn the manifold depicted in , all points are represented by the same shape with the same color. Therefore, in Figure 4B In a manifold of , the only discernible information provided by a point is its position (x,y) on the manifold. However, in Figure 4B It is not possible to distinguish between the individual gene transcript counts and the specific day on which the gene transcript counts for each point were obtained. As discussed in further detail below, Figure 5A The shapes of the points in the manifold of vary to indicate, in part, the day on which the gene transcript counts for each point, and therefore each cell, were obtained. Similarly, Figure 5B The shading of the points in the manifold is varied to indicate the gene transcript counts on a gene-by-gene basis for each point and therefore each cell on each measurement day.

[0361] Figure 5A Describes an embodiment of Figure 4B In the manifold. Figure 5A In the embodiment of the manifold depicted in , each point in the manifold is labeled with the day on which the transcription factor expression of the cell associated with the point was measured and the qualitative stage in the transformation process that the cell is in. For example, Figure 5A Points marked with squares in the manifold in indicate that the points are associated with day 5 cells, which are qualitatively characterized as early induced neuronal (iN) cells.

[0362] By labeling each point in the manifold with the day on which gene expression of the cell associated with that point was measured and the qualitative stage of cellular transformation, transition trajectories can be identified. Figure 5A The arrows below the manifolds in Figure 1 indicate an identified trajectory depicting the transformation of MEF cells into mouse neurons. Figure 5A Another identified trajectory in the image depicts the transformation of MEF cells into mouse myocytes.

[0363] By identifying the difference in gene expression between points (e.g., cells) at different stages along the transition track, it is possible to identify genes that contribute to the transition of cells along a specific track. But perhaps more importantly, by identifying the difference in gene expression between points (e.g., cells) at the junctions where two or more transition tracks diverge, it is possible to identify genes that contribute to the divergence in the transition track. These genes identified can then be predicted to be associated with specific transition tracks and / or stages. For example, if the expression level of gene A is identified in cells labeled as early iN cells on the 5th day relative to cells labeled as early myocytes on the 5th day, it can be assumed that the expression of gene A is associated with the transition track from MEF to mouse neurons, and vice versa with the transition track from MEF to mouse myocytes.

[0364] As discussed above, Figure 5AThe transformation trajectory was established based on both the quantitative time points during the cell transformation process and the qualitative stages of the cell transformation process. Figure 5A There is no indication of the level of gene expression on a gene-by-gene basis at each point (e.g., cells at different time points). Figure 5A Based on the information depicted in , it is not possible to predict which genes are associated with which transition trajectories. However, as mentioned above, Figure 5B The shading of the points in the manifold changes to indicate the relative gene transcript counts at each point on a gene-by-gene basis. Based on this gene-by-gene description of gene expression at these points (e.g., cells at different time points), it is possible to predict which genes are associated with which transition trajectories.

[0365] Figure 5B Depicting a depiction according to one embodiment is Figure 4B The expression level of each of the three BAM transcription factors in each cell at each measurement day (day 2, day 5, and day 22 for Ascl1, and day 22 for Brn2 and Myt11) for each point in the manifold of . Specifically, Figure 5B Depiction Figure 4B Three different forms of manifolds. Figure 5B The first form of the manifold depicted in depicts the expression level of the Ascl1 transcription factor at each point of the manifold, Figure 5B The second form of the manifold depicted in , depicts the expression level of the Brn2 transcription factor at each point of the manifold, and Figure 5B The third form of the manifold depicted in FIG. 4 depicts the expression level of the Myt1l transcription factor at each point of the manifold.

[0366] exist Figure 5B In , the expression level of a transcription factor at a point in the manifold (e.g., a cell at a time point) is measured as the logarithm of fragments per kilobase of transcript per million mapped reads (FPKM) of the transcription factor. A relatively low log(FPKM) value indicates a relatively low transcription factor expression level. On the other hand, a relatively high log(FPKM) value indicates a relatively high transcription factor expression level. In Figure 5B In the manifold of , a relatively low transcription factor expression level (e.g., a relatively low log(FPKM) value) of the point is indicated by making the shade of the point relatively darker. Conversely, a relatively high transcription factor expression level (e.g., a relatively high log(FPKM) value) of the point is indicated by making the shade of the point relatively lighter.

[0367] By comparison Figure 5A The transformation trajectory depicted in Figure 5B A manifold depicting the gene-by-gene expression levels of BAM transcription factors identifies transcription factors that influence the progression of cells along specific transformation trajectories.

[0368] Turning first to the graph depicting the expression of the Ascl1 transcription factor at day 0 during the 23 day period Figure 5B In the manifold, mouse cells were transduced with either Ascl1 or BAM alone. Thus, day 0 cells did not express Ascl1 at detectable levels. These day 0 cells that did not express Ascl1 were MEFs. Then, on day 2 of the 23-day period, Ascl1 was expressed at relatively low levels, as depicted by the relatively dark shading of the dots associated with day 2 cells. These day 2 cells expressing Ascl1 began to follow Figure 5A Specifically, some day 2 cells become mouse progenitors, some day 2 cells become intermediate cells on the MEF-to-neuron trajectory, and some day 2 cells become induced cells on the MEF-to-myocyte trajectory. Similarly, on day 5 of the 23-day time period, expression of Ascl1 in day 5 cells increases relative to day 2 cells, as depicted by the relatively lighter shading of the dots associated with day 5 cells. These day 5 cells with increased Ascl1 expression further along the Figure 5A Specifically, day 5 cells on the MEF-to-neuron trajectory became intermediate and early iN cells, whereas day 5 cells on the MEF-to-myocyte trajectory became early myocytes. Finally, at day 22 of the 23-day period, Ascl1 expression in day 22 cells increased or remained the same relative to day 5 cells. These day 22 cells expressing Ascl1 further progressed along the Figure 5A Specifically, day 22 cells on the MEF-to-neuron trajectory became mature mouse neurons, while day 22 cells on the MEF-to-myocyte trajectory became mature mouse myocytes. No mouse progenitor cells remained at day 22.

[0369] These observations on MEF cell state transitions following induction of Ascl1 expression follow trends known in the literature. Specifically, as briefly discussed above, Ascl1 priming induces MEFs to transform into mouse progenitor cells, and Ascl1 expression alone induces mouse progenitor cells to transform into mouse neurons and mouse myocytes. Figure 5B As discussed in detail in the Ascl1 manifolds, after forced expression of Ascl1 in day 0 MEFs, MEFs transitioned into any of mouse progenitor cells, mouse myocytes, and mouse neurons.

[0370] Next, we turn to characterizing the expression of the Brn2 transcription factor Figure 5B In the manifold, MEFs were transduced with BAM factors on day 0 of the 23-day time period. Brn2 expression was measured only on day 22 during the 23-day time period. Figure 5BAs shown in Figure 2, at day 22 of a 23-day period, mouse neurons strongly expressed Brn2 on day 22. Therefore, it can be inferred that the expression of Brn2 is associated with the progression of MEF cells along the transformation trajectory of MEFs to mouse neurons.

[0371] This observation of MEF cell state transition following induction of Brn2 expression follows trends known in the literature. Specifically, as briefly discussed above, Brn2 expression induces the transformation of mouse progenitor cells into mouse neurons. Figure 5B As discussed in the Brn2 manifold, MEFs expressing Brn2 were transformed into mouse neurons.

[0372] Finally, we turn to the description of the expression of the Myt1l transcription factor Figure 5B In the manifold, MEFs were transduced with BAM factors on day 0 of the 23-day timeframe. Myt1l expression was measured only on day 22 during the 23-day timeframe. Myt1l was strongly expressed by day 22 mouse neurons on day 22 of the 23-day timeframe. Therefore, similar to the Brn2 transcription factor, it can be inferred that Myt1l expression is associated with the progression of MEF cells along the transformation trajectory from MEFs to mouse neurons.

[0373] This observation of a transition in MEF cell states following induction of Myt11 expression follows trends known in the literature. Specifically, as briefly discussed above, Myt11 expression induces the transition of mouse progenitor cells into mouse neurons. Figure 5B As discussed in the Myt1l manifold, MEFs expressing Myt1l were transformed into mouse neurons.

[0374] Therefore, by using the methods of Part II and Part III to generate Figure 5B These observations obtained with the Ascl1, Brn2, and Myt1l manifolds in the literature are consistent with observations documented in the literature. This consistency in the observations of Ascl1-assisted transitions, Brn2-assisted transitions, and Myt1l-assisted transitions helps validate the ability of the methods in Parts II and III to accurately identify genes that affect cell state transitions.

[0375] To further validate the ability of the methods in Parts II and III to accurately identify genes that affect cell state transitions, in vitro experiments were performed to confirm the ability of the methods in Parts II and III to accurately identify genes that affect cell state transitions. Figure 5A and Figure 5B Specifically, in vitro experiments were performed to confirm the above observations, namely that Ascl1 expression induces MEFs to transform into mouse "progenitor" cells, mouse neurons and / or mouse muscle cells.

[0376] The in vitro experiment was performed according to the protocol listed in Section IV.A above. As discussed above, in the protocol, only the expression of Ascl1 was forced in MEFs. After the expression of the Ascl1 transcription factor was forced in MEFs on day 0 of the 23-day period, the mouse cells were stained with DAPI, Map2 antibody, and Tuj1 antibody on day 15 of the 23-day period. It is known that DAPI stains regions rich in adenine-thymine in DNA. Therefore, DAPI stains the nucleus. It is known that Map2 antibody and Tuj1 antibody stain neural cells. Therefore, by staining mouse cells with DAPI, Map2 antibody, and Tuj1 antibody, the amount of mouse neurons relative to the amount of overall mouse cells can be identified, and therefore the effect of Ascl1 overexpression on MEF transformation can be determined. In the in vitro experiment, this group of mouse cells in which the Ascl1 transcription factor is forced to be expressed is referred to as the experimental group in this article.

[0377] As a positive control group in the in vitro experiment, samples of mouse cells including only mouse neurons were also stained with DAPI, Map2 antibody, and Tuj1 antibody. As a negative control group, samples of MEF cells without forced Ascl1 expression were also stained with DAPI, Map2 antibody, and Tuj1 antibody.

[0378] After staining the experimental group, positive control group, and negative control group with DAPI, Map2 antibody, and Tuj1 antibody, each group stained with each dye was imaged on a Molecular Devices HCI IXM4. The resulting images are shown in Figure 6 middle. Figure 6 Depicting images of MEF cells with forced Ascl1 expression that have been stained with DAPI, Map2 antibody, and Tuj1 antibody, images of mouse neurons stained with DAPI, Map2 antibody, and Tuj1 antibody, and images of MEF cells without forced Ascl1 expression that have been stained with DAPI, Map2 antibody, and Tuj1 antibody according to one embodiment.

[0379] First, we turn to the images of the negative control group, such as Figure 6 As shown in , nuclei of DAPI-stained MEF cells without forced Ascl1 expression are visible, but there are almost no neurons in the images depicting Map2 and Tuj1 staining of MEF cells without forced Ascl1 expression. In other words, although there are many mouse cells (especially MEFs) in the sample, there are no neurons. This is an expected result because there is no forced Ascl1 expression in the MEF cells of the sample, and therefore the transformation of MEF cells into neurons is not induced.

[0380] Next, we turn to the images of the positive control group, such as Figure 6As shown in , the nuclei of DAPI-stained mouse neurons are visible, and these same mouse neurons are also visible in images depicting Map2 and Tuj1 staining of mouse neurons. In other words, all cells in the positive control sample were accurately identified as neurons.

[0381] Finally, we turn to the images of the experimental group, e.g. Figure 6 As shown in , the nuclei of DAPI-stained MEF cells with forced Ascl1 expression are visible. In addition, some of these DAPI-stained cells are also stained with Map2 and Tuj1, indicating that these selected cells are mouse neurons. Therefore, it can be inferred that forced expression of Ascl1 is associated with the induction of the transformation of MEFs into mouse neurons.

[0382] Figure 6 In vitro experiments confirmed that forced expression of Ascl1 in MEF cells can lead to the transformation of MEF cells into mouse neurons, as described above. Figure 5A and Figure 5B Observed in the computer experiments described. Figure 5A and Figure 5B This confirmation of the observations in further validates the ability of the methods of Parts II and III to accurately identify genes that affect cell state transitions.

[0383] IV.D. Example 2: Clustering

[0384] As discussed above in Section III.C., after generating the matrix M by dimensionality reduction, clustering is performed to group the data in the matrix M to generate clusters C j Cluster C j Each cluster in the set consists of a set of points.

[0385] Fig. 7A Describes an embodiment of Figure 4B manifold where the points in the manifold are grouped into clusters C identified by clustering j In. Fig. 7A In an embodiment of the present invention, clustering is performed using Louvain community detection, in particular GenLouvain community detection. Fig. 7A As can be seen in Figure 2, clustering identifies 10 unique clusters C of points in the manifold. j .

[0386] Typically, clustering assigns points in a manifold to a given cluster based on a threshold similarity of values ​​associated with the points, e.g., their location in the reduced dimensionality space of the manifold, their associated gene transcript counts, etc. Specifically, for Fig. 7A Manifolds of , clustering assigns points to given clusters based on a threshold similarity between points in the manifold. For example, Fig. 7A The points in the manifold of that included in group 8 are likely to be associated with mouse neurons or other cells that are genetically similar to mouse neurons. Similarly, Fig. 7A The points in the manifold included in group 9 are all likely associated with mouse myocytes or other cells genetically similar to mouse myocytes.

[0387] As discussed above, in addition to being able to accurately identify genes known in the literature to induce cell state transitions, the methods of Parts II and III also allow for the identification of factors (e.g., genes and perturbations) that are not known in the literature to affect cell state transitions. Figure 7B Describe the transcription factors associated with the transformation of MEFs to mouse neurons (and vice versa) to mouse myocytes, both known and unknown in the literature, according to one embodiment. In particular, Figure 7B Describe the transcription factors associated with inhibiting the transition of mouse "progenitor" cells to mouse muscle cells when underexpressed in mouse "progenitor" cells, and the transcription factors associated with the transition of mouse "progenitor" cells to mouse neurons when overexpressed in mouse "progenitor" cells. By underexpressing the transcription factors associated with inhibiting the transition of mouse "progenitor" cells to mouse muscle cells in mouse "progenitor" cells, and by overexpressing the transcription factors associated with inducing the transition of mouse "progenitor" cells to mouse neurons in mouse "progenitor" cells, the transition of mouse "progenitor" cells to mouse neurons (and vice versa mouse muscle cells) can be induced.

[0388] In order to identify transcription factors associated with a transition from a first cell state to an alternative specific cell state or from a first cell state to any other cell state, clustering can be used. Specifically, gene transcript counts associated with points in a cluster associated with the first cell state are identified and compared with gene transcript counts associated with points in another cluster associated with an alternative specific cell state or with any cell state other than the first cell state. This comparison of gene transcript counts between clusters can be performed using any differential expression test (e.g., mean difference test, Wilcoxon rank sum test, t-test, logistic regression, and generalized linear model).

[0389] As an example, to identify transcription factors associated with the transition from MEFs to mouse neurons, we used Fig. 7A First, to identify transcription factors that, when overexpressed in mouse progenitor cells, are associated with the transition of mouse progenitor cells to mouse neurons, Fig. 7A of clusters associated with mouse neurons (e.g., Fig. 7A The gene transcript counts associated with the points included in cluster 8) of Fig. 7A Transcript counts of genes associated with spots included in an alternative cluster not associated with mouse neurons were compared. Figure 7B In an embodiment of the present invention, the Wilcoxon rank sum test is used to perform this comparison. However, in alternative embodiments, any other statistical analysis method can be used to perform the comparison. Based on this comparison, it is predicted that Fig. 7A The genes overexpressed in cells associated with the dots in the cluster associated with mouse neurons are associated with the transition of mouse "progenitor" cells to mouse neurons. The transcription factors produced by the transcription and translation of these genes are identified as Figure 7B A transcription factor that, when overexpressed in mouse progenitor cells, is associated with the transition of mouse progenitor cells to mouse neurons.

[0390] Similarly, to identify transcription factors that, when lowly expressed in mouse progenitor cells, are associated with the inhibition of the transition of mouse progenitor cells to mouse myocytes, Fig. 7A of clusters associated with mouse myocytes (e.g., Fig. 7A The gene transcript counts associated with the points included in cluster 9) of Fig. 7A The transcript counts of genes associated with the spots included in the alternative cluster not associated with mouse myocytes were compared. Figure 7B In an embodiment, the Wilcoxon rank sum test is used to perform this comparison. However, in alternative embodiments, any other statistical analysis method can be used to perform the comparison. Based on this comparison, it is predicted that Fig. 7A The genes that are under-expressed in cells associated with the points in the cluster associated with mouse myocytes are associated with inhibiting the transition of mouse "progenitor" cells to mouse myocytes. The transcription factors produced by the transcription and translation of these genes are identified as Figure 7B A transcription factor that, when lowly expressed in mouse progenitor cells, is associated with the inhibition of the transition of mouse progenitor cells to mouse muscle cells.

[0391] like Figure 7B As shown in, when overexpressed in mouse "progenitor" cells, the transcription factors associated with the transformation of mouse "progenitor" cells to mouse neurons include Zfp941, Brn2, Myt1l, Taf5B, St18, Zkscan16, Camta1 and Arnt2. When low-expressed in mouse "progenitor" cells, the transcription factors associated with inhibiting the transformation of mouse "progenitor" cells to mouse muscle cells include Atf3, Rorc, Scx, Satb1, Elf3 and Fos. As discussed in detail above with respect to Example 1, it is known in the literature that Brn2 and Myt1l transcription factors are associated with inducing the transformation of mouse "progenitor" cells to mouse neurons. However, it is not known in the literature that Figure 7BThe remaining transcription factors described in have been associated with the transition of MEFs to mouse neurons (and vice versa to mouse myocytes). Thus, by using the methods described in Parts II and III above, genes and / or transcription factors known and unknown in the literature to induce cells to follow a specific transition trajectory can be identified. These identified transcription factors can then be used to control cell state transitions and, therefore, cell fate.

[0392] IV.E. Example 3: Perturbation-Induced Transformation

[0393] As discussed in Sections III.D and III.E, in addition to being able to identify genes and transcription factors that affect cell state transitions, the methods of Sections II and III are also able to identify perturbations, such as small molecules, that affect cell state transitions. First, in order to identify perturbations that induce cells to follow a specific transition trajectory, possible transition trajectories are identified.

[0394] Fig. 8A Describes a method for Figure 4A Map of the transition trajectory of the MEF cells in question. To construct this map of transition trajectories, we used Figure 4B Specifically, points in the manifold associated with similar gene transcript counts are grouped into states (in Fig. 8A Points with variable gene transcript counts between states are used to identify transition paths between states (in Fig. 8A ). You can use Fig. 8A The map of transition trajectories depicted in the figure identifies perturbations that affect the transition trajectory of a cell by changing gene expression in the cell and thereby causing the cell to progress from one state in the map of transition trajectories to another state. In some embodiments, to generate Fig. 8A The mapping of transition trajectories depicted in can be used to classify cells via a set of canonical marker genes. In such an embodiment, cells identified as the same cell type are predicted to be cells along the same transition trajectory in the mapping of transition trajectories. In an alternative embodiment, to generate Fig. 8A Mapping the transformation trajectories depicted in Figure 4B The branches of the manifold are used to define different transition trajectories in the mapping of transition trajectories.

[0395] Figure 8B An example of a method described in Section III.D. for identifying a perturbation that affects a transition trajectory of a cell by altering gene expression in the cell such that the cell undergoes a transition in accordance with an embodiment of the present disclosure. Fig. 8ASpecifically, in order to identify a perturbation that causes a cell to change gene expression when exposed to the cell so that the cell changes from the first state to the second state, Figure 8B The method compares the change in gene expression in the cell after the cell transitions from the first state to the second state with the change in gene expression in the vehicle cell after the vehicle cell is exposed to the perturbation. If the change in gene expression in the cell after the cell transitions from the first state to the second state matches (e.g., is equivalent or similar to) the change in gene expression in the vehicle cell after the vehicle cell is exposed to the perturbation, it can be predicted that the perturbation induces the cell exposed to the perturbation to transition from the first state to the second state by changing gene expression in the cell. In this way, it can be predicted that the perturbation is associated with a specific trajectory of cell state transitions.

[0396] Specifically turn to Figure 8B In the embodiment depicted in Figure 8B Depicted are gene expression levels of six different genes (Genes 1 to 6) for cells in State 1, cells in State 2, vehicle cells, and vehicle cells exposed to small molecule perturbations. The gene expression level of a given gene is depicted by shading. Polka-dot shading indicates undetectable gene expression, while cross-hatching shading indicates detectable gene expression. In other words, in Figure 8B In some embodiments, gene expression is measured on a binary basis (detectable gene expression or undetectable gene expression). However, in alternative embodiments, gene expression levels are not measured on a binary basis, but on a more quantitative basis.

[0397] Turning to examining the gene expression levels of each gene in each cell, for cells in state 1, expression of genes 1 to 3 was undetectable, but expression of genes 4 to 6 was detectable. Conversely, for cells in state 2, expression of genes 4 to 6 was undetectable, but expression of genes 1 to 3 was detectable. For vehicle cells, expression of genes 1 to 3 was undetectable, but expression of genes 4 to 6 was detectable. Conversely, for vehicle cells exposed to a perturbation, expression of genes 4 to 6 was undetectable, but expression of genes 1 to 3 was detectable.

[0398] Next, for each gene, the gene expression level in cells in state 1 is compared to the gene expression level in cells in state 2 to determine the change in gene expression level after the cell transitions from state 1 to state 2. As indicated by the dark cross-hatched shading associated with genes 1 to 3, the expression of genes 1 to 3 increases after the cell transitions from state 1 to state 2. On the other hand, as indicated by the dark polka-dot shading associated with genes 4 to 6, the expression of genes 4 to 6 decreases after the cell transitions from state 1 to state 2.

[0399] Similarly, for each gene, the gene expression level in the vehicle cells was compared to the gene expression level in the vehicle cells exposed to the perturbation to determine the change in gene expression level after the vehicle cells were exposed to the perturbation. As indicated by the dark cross-hatched shading associated with genes 1 to 3, the expression of genes 1 to 3 increased after the vehicle cells were exposed to the perturbation. On the other hand, as indicated by the dark polka-dot shading associated with genes 4 to 6, the expression of genes 4 to 6 decreased after the vehicle cells were exposed to the perturbation.

[0400] Finally, the change in gene expression in the cell after the cell is converted from state 1 to state 2 is compared with the change in gene expression in the vehicle cell after the vehicle cell is exposed to the disturbance. In order to compare the change in gene expression in the converted cell with the change in gene expression in the vehicle cell, any differential expression test can be used. For example, any of the mean difference test, Wilcoxon rank sum test, t test, logistic regression and generalized linear model comparison algorithms can be used.

[0401] like Figure 8B As shown in , the expression of genes 1 to 3 increased in cells that transitioned from state 1 to state 2 and in the vehicle cells exposed to the perturbation. In addition, the expression of genes 4 to 6 decreased in cells that transitioned from state 1 to state 2 and in the vehicle cells exposed to the perturbation. Based on this similarity in the changes in gene expression in cells that transitioned from state 1 to state 2 and in the vehicle cells exposed to the perturbation, it can be predicted that by changing the gene expression in the cells, exposure of cells in state 1 to perturbations can induce cells in state 1 to transition to state 2. Therefore, perturbations can be used to control the transition of cells from state 1 to state 2.

[0402] Above about Figure 8B The method described involves the identification of a perturbation associated with inducing a cell to transition from a general state 1 to a general state 2. Figure 8B The described method can be used to identify induced cells from Fig. 8A The transition trajectory maps any state into any other state associated with a perturbation. However, instead of referring to Fig. 8AThe general state in the mapping of the transformation trajectory, Fig. 9 Identification Fig. 8A Mapping of transition trajectories in specific states and then identifying cells associated with induction or inhibition of Fig. 9 The specific perturbation associated with the transition from one recognition state to another recognition state causes the cell to become a mouse neuron (or vice versa). Fig. 9 Identify a MEF state, a mouse "progenitor" cell state, a mouse myocyte state, and a mouse neuron state, and then identify specific perturbations associated with inducing or inhibiting the transition of a cell from one of these states to another, such that the cell becomes a mouse neuron (or vice versa).

[0403] Fig. 9 Describes small molecule perturbations associated with the transition of MEFs to mouse neurons (and vice versa to mouse myocytes) according to one embodiment. In particular, Fig. 9 Describe the set of small molecule perturbations associated with the transformation of MEF to mouse "progenitor" cells when exposed to MEF, the set of small molecule perturbations associated with the transformation of mouse "progenitor" cells to mouse muscle cells when exposed to mouse "progenitor" cells, and the small molecule perturbations associated with the transformation of mouse "progenitor" cells to mouse neurons when exposed to mouse "progenitor" cells. MEFs can be induced to transform into mouse neurons (and vice versa mouse muscle cells) by exposing MEFs to perturbations associated with inducing MEF to transform into mouse neurons, exposing MEFs to perturbations associated with inhibiting the transformation of mouse "progenitor" cells to mouse muscle cells, and exposing MEFs to perturbations associated with inducing mouse "progenitor" cells to transform into mouse neurons.

[0404] By implementing the above Figure 8B The method described identifies Fig. 9 For example, to identify the small molecule perturbation BRD-K38615104 that is associated with the transition of MEFs to mouse "progenitor" cells, use Figure 8B The method determines that the changes in gene expression in MEFs after the MEFs are transformed into mouse "progenitor" cells match (e.g., are equivalent or similar) the changes in gene expression in vehicle cells after the vehicle cells are exposed to BRD-K38615104. And therefore, it is predicted that BRD-K38615104 induces the transformation of MEFs into mouse "progenitor" cells by changing gene expression in MEFs. Similarly, to identify the small molecule perturbation Dasatinib associated with the inhibition of the transformation of mouse "progenitor" cells into mouse muscle cells, use Figure 8BThe method determined that the changes in gene expression in mouse "progenitor" cells after the transition of mouse "progenitor" cells to mouse muscle cells were the reversal of the changes in gene expression in vehicle cells after vehicle cells were exposed to dasatinib. And therefore, it is predicted that dasatinib inhibits the transition of mouse "progenitor" cells to mouse muscle cells.

[0405] like Fig. 9 As seen in the , small molecule perturbations associated with the transition of MEFs to mouse "progenitor" cells when exposed to MEFs include BRD-K38615104, geldanamycin, manumycin A, mitoxantrone, curcumin, and trichostatin A. Small molecule perturbations associated with the transition of mouse "progenitor" cells to mouse neurons when exposed to mouse "progenitor" cells include avosidide, varinostat, KI20227, forskolin, PP1, and PP2. Small molecule perturbations associated with inhibiting the transition of mouse "progenitor" cells to mouse muscle cells when exposed to mouse "progenitor" cells include avosidide, geldanamycin, quinacrine, CGP-60474, and dasatinib.

[0406] Fig. 9 Two of the small molecule perturbations identified in the study, avosidix and geldanamycin, have been implicated in inducing the transition of mouse cells into mouse neurons by inducing and / or inhibiting the transition of mouse cells into two distinct states. Fig. 9 As shown in , avosidic acid is associated with both inducing the transformation of mouse "progenitor" cells into mouse neurons and inhibiting the transformation of mouse "progenitor" cells into mouse muscle cells. Similarly, geldanamycin is associated with both inducing the transformation of MEFs into mouse "progenitor" cells and inhibiting the transformation of mouse "progenitor" cells into mouse muscle cells. Therefore, by exposing MEFs to avosidic acid and geldanamycin, it can be predicted that MEFs will transform into mouse neurons.

[0407] exist Fig. 9 Some of the small molecule perturbations identified in are known in the literature to be associated with the indicated transformation trajectories. Specifically, forskolin, PP1, and PP2 are known in the literature to be associated with inducing the transformation of mouse "progenitor" cells to mouse neurons. Similarly, trichostatin A is known in the literature to be associated with inducing the transformation of MEFs to mouse "progenitor" cells. Figure 8B This agreement between the predictions made by the method and the information known in the literature indicates that Figure 8B The ability of our method to accurately identify perturbations that affect cell-state transitions.

[0408] In addition to accurately identifying perturbations known in the literature to affect cell state transitions, Figure 8B The method can also identify perturbations that are not known in the literature to affect cell state transitions. Fig. 9The remaining small molecule perturbations described in were associated with the transition of MEFs to mouse neurons (and vice versa). Figure 8B The described method allows identification of perturbations known and unknown in the literature that induce cells to follow specific transition trajectories. These identified perturbations can then be used to control cell state transitions and, therefore, cell fate.

[0409] V. Example 4

[0410] The experiments of this example demonstrate methods for promoting neurons and / or progenitor cells. In the experiments described herein, a population of starting fibroblasts (i.e., primary mouse fibroblasts) was exposed to a composition comprising an Ascl1 overexpressing lentivirus. After 48 hours, a compound (e.g., forskolin, gesartinib, PD-0325901) or a vehicle (i.e., DMSO or ethanol) was added to the composition. The total number of neurons was manually counted based on positive Tuj1 / Map2 signals and neuronal morphology. For each experiment, the total number of neurons for each treatment condition was normalized by the number of neurons in the DMSO-treated wells relative to the number of neurons in the experiment. As Fig. 10A and 10B As shown in, the presence of neurons developed from the starting fibroblast population was detected in these experiments. Depending on the compound added to the composition, the fold change of both the total number of neurons and the percentage of neurons increased, decreased or remained the same. These experiments show that the methods of the present invention can be used to promote neurons and / or progenitor cells from a starting cell population including fibroblasts.

[0411] Cell culture and compound treatment

[0412] Primary mouse embryonic fibroblasts (MEFs) at passage 2 were plated at 20,000-45,000 / well (depending on batch) in 24-well plates in MEF medium containing 10% FBS, 1x Glutamax, 1x MEM non-essential amino acids, 1mM sodium pyruvate, 0.05U / ml Pen / Strep and 55μM β-mercaptoethanol in DMEM. After 24 hours of culture, MEFs were infected with Ascl1 overexpressing lentivirus in MEF medium containing 8μg / ml polybrene by centrifugal transfection (plates were centrifuged at 2000rpm at 32°C for 90 minutes). See below for lentivirus generation. After 48 hours, the medium was replaced with neuronal medium containing DMEM / F12, 1% N2, 2% B27 1:50, 1x Glutamax, 25μg / ml insulin, 0.05U / ml Pen / Strep containing compounds or vehicles (DMSO or ethanol). The compounds and their concentrations are selected from the following: BI-2536 (200 nM), Cilostazol (1000 nM), Dabrafenib (2500 nM), Estradiol-Cypionate (2000 nM), EX-527 (5000 nM), Fedratinib (1000 nM), Foretinib (200 nM), Forskolin (5000 nM), Glesatinib (2500 nM), Indirubin 3 Oxime (2000 nM), KI20227 (250 nM), KU 0060648 (200 nM), m-3M3FBS (1000 nM), manumycin (800 nM), PD-0325901 (5000 nM), PHA-665752 (1000 nM), Quinacrine (200 nM), Rottlerin (1000 nM), Selumetinib (100 nM), Troglitazone (5000 nM) and Vemurafenib (5000 nM). Perform a half medium change with supplemented compounds every 2 to 3 days.

[0413] Immunofluorescence staining

[0414] At day 12 after Ascl1 infection, cells were fixed with 4% paraformaldehyde, permeabilized (0.2% Triton X100) and blocked in 5% serum (mixture of donkey, calf, and goat serum), and stained with rabbit anti-Tuj1 (1:1000) and mouse anti-Map2 (1:500) antibodies overnight at 4°C or at room temperature for 2 h, followed by secondary antibody and DAPI staining.

[0415] Imaging and analysis

[0416] Imaging was performed on a Molecular Devices ImageXpress Micro; 36 images per well were acquired from a 10x objective. The total number of neurons was manually counted based on positive Tuj1 / Map2 signals and neuronal morphology. For each experiment, the total number of neurons for each treatment condition was normalized by the number of neurons in the DMSO-treated wells for that experiment.

[0417] Lentivirus generation

[0418] Lentivirus was packaged by transfecting 293T cells with packaging plasmid (SystemsBio, LV510A-1) or analog and Ascl1 overexpression plasmid (Ascl1 cDNA cloned into Origene lentiviral expression vector accession number PS100064) via Mirus TransITLenti transfection reagent (Mirus, MIR 6603) and concentrated in a BeckmanCoulter ultracentrifuge at 16,500RPM for 1.5 hours. Only experiments were performed in which 90% or more of the cells were infected with lentivirus, as judged by immunofluorescence staining with rabbit anti-Ascl1 (1:200; Abcam, ab74065-100UG) for 48 hours.

[0419] V. Example 5

[0420] Embodiment 1. A method for predicting whether a perturbation will affect a cell transition, the method comprising: on a computer system comprising a memory and one or more processors: accessing in electronic form a single cell transition signature, the single cell transition signature representing a measure of differential cellular component expression between a first cell state and an altered cell state, wherein the altered cell state occurs by a cell transition from the first cell state to the altered cell state, and wherein the single cell transition signature comprises identification of a plurality of cellular components and, for each respective cellular component of the plurality of cellular components, quantifying a correspondence of an association between a change in expression of the respective cellular component and a change in cellular state between the first cell state and the altered cell state a first significance score; accessing a perturbation signature in electronic form, the perturbation signature representing a measure of differential cellular component expression between a plurality of unperturbed cells and a plurality of perturbed cells exposed to the perturbation, wherein the perturbation signature comprises an identification of all or a portion of the plurality of cellular components and, for each respective cellular component in the all or a portion of the plurality of cellular components, a corresponding second significance score quantifying an association between (i) a change in expression of the respective cellular component between the plurality of unperturbed cells and the plurality of perturbed cells and (ii) a change in cellular state between the plurality of unperturbed cells and the plurality of perturbed cells; and comparing the single cell transition signature and the perturbation signature to determine whether the perturbation will affect the cellular transition.

[0421] Embodiment 2. A method as described in embodiment 1, wherein accessing the single-cell transition feature includes: determining the single-cell transition feature based on (i) a first plurality of first single-cell cell component expression data sets and (ii) a second plurality of second single-cell cell component expression data sets, wherein: each corresponding first single-cell cell component expression data set in the first plurality of first single-cell cell component expression data sets is obtained from a corresponding single cell of a first plurality of cells in the first cell state, and each corresponding second single-cell cell component expression data set in the second plurality of second single-cell cell component expression data sets is obtained from a corresponding single cell of a second plurality of cells in the changed cell state.

[0422] Embodiment 3. A method as described in Embodiment 2, wherein: each corresponding data set of the first plurality of single-cell cellular component expression data sets includes a corresponding cellular component vector in the first plurality of cellular component vectors, each corresponding data set of the second plurality of single-cell cellular component expression data sets includes a corresponding cellular component vector in the second plurality of cellular component vectors, each corresponding cellular component vector in the first plurality of cellular component vectors and the second plurality of cellular component vectors contains a plurality of elements, each corresponding element in the corresponding cellular component vectors is associated with a corresponding cellular component in the plurality of cellular components and includes a corresponding value representing the amount of the corresponding cellular component of the corresponding single cell, and the corresponding single cell is represented by the corresponding data sets of the first plurality of single-cell cellular component expression data sets and the second plurality of single-cell cellular component expression data sets.

[0423] Embodiment 4. A method as described in Embodiment 3, which also includes: performing dimensionality reduction on the first plurality of single-cell cell component expression data sets and / or the second plurality of single-cell cell component expression data sets to generate a plurality of reduced-dimensionality components; for each corresponding cell component vector in the first plurality of cell component vectors and the second plurality of cell component vectors, applying the plurality of reduced-dimensionality components to the corresponding cell component vector to form a corresponding reduced-dimensionality vector, the corresponding reduced-dimensionality vector including the reduced-dimensionality component value of each corresponding reduced-dimensionality component in the plurality of reduced-dimensionality components, thereby forming a corresponding first plurality of reduced-dimensionality vectors and a second plurality of reduced-dimensionality vectors; and performing clustering to generate clusters C j A set of clusters C, each cluster containing a plurality of points corresponding to a subset of the first plurality of reduced dimensionality vectors and the second plurality of reduced dimensionality vectors; j a first cluster of the set of cells identifying the first plurality of cells; and j A second cluster of the set identifies the second plurality of cells, the method optionally further comprising performing manifold learning with the corresponding first plurality of reduced dimensionality vectors and second plurality of reduced dimensionality vectors to identify a relative cell state of each cell in the first plurality of cells and the second plurality of cells relative to each other cell.

[0424] Embodiment 5. A method as described in any of embodiments 1 to 4, wherein the plurality of unperturbed cells are control cells that have not been exposed to the perturbation, or wherein the unperturbed cells are an average of unrelated disturbed cells that have been exposed to the perturbation.

[0425] Embodiment 6. A method as described in any one of Embodiments 1 to 5, further comprising: pruning the single-cell transition signature and the perturbation signature to limit the multiple cellular components to transcription factors, optionally measured at the RNA level.

[0426] Embodiment 7. A method as described in Embodiment 2, wherein the determining the single-cell transformation characteristics includes: using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model to determine the differences in the cellular component quantities of the multiple cellular components between (i) the first plurality of first single-cell cellular component expression data sets and the second plurality of second single-cell cellular component expression data sets.

[0427] Embodiment 8. A method as described in embodiment 1, wherein the measure of differential cellular component expression uses one of a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model to quantify the difference in cellular component amount between (i) a third plurality of third single-cell cellular component expression datasets and (ii) a fourth plurality of fourth single-cell cellular component expression datasets, wherein: each corresponding third single-cell cellular component expression dataset in the third plurality of third single-cell cellular component expression datasets is obtained from a corresponding single cell in the plurality of unperturbed cells, and each corresponding fourth single-cell cellular component expression dataset in the fourth plurality of fourth single-cell cellular component expression datasets is obtained from a corresponding single cell in the fourth plurality of perturbed cells exposed to the perturbation.

[0428] Embodiment 9. A method as described in any one of Embodiments 1 to 8, which also includes: filtering the single-cell transition feature and the perturbation feature to reduce the number of cellular components included in the single-cell transition feature and the perturbation feature, optionally wherein the resulting filtering of the single-cell transition feature and the perturbation feature includes reducing the number of cellular components included in the single-cell transition feature and the perturbation feature according to a threshold p value or according to a threshold number of cellular components.

[0429] Embodiment 10. A method as described in any one of embodiments 1 to 9, wherein determining the corresponding second significance score of the corresponding cellular component comprises: for each corresponding cellular component in the multiple cellular components, replacing the significance score of the corresponding cellular component with the corresponding matching score of the corresponding cellular component; combining the matching scores of the multiple cellular components to generate a matching score for the perturbation; and determining whether the corresponding perturbation is associated with a transition of the cell between the first cellular state and the changed cellular state based on the matching score of the perturbation, optionally wherein the corresponding matching score comprises a discrete or continuous score.

[0430] Embodiment 11. A method as described in embodiment 10, wherein replacing the significant score includes: if the amount of the cellular component of the corresponding cellular component from the single-cell transition feature and the amount of the cellular component of the corresponding cellular component from the perturbation feature are both upregulated, replacing the significant score with a first score; if the amount of the cellular component of the corresponding cellular component from the single-cell transition feature is upregulated and the amount of the cellular component of the corresponding cellular component from the perturbation feature is downregulated, replacing the significant score with a second score; and if the amount of the cellular component of the corresponding cellular component from the perturbation feature is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0431] Embodiment 12. A method as described in embodiment 10, wherein replacing the significant score includes: if the amount of the cellular component of the corresponding cellular component from the single-cell transition feature and the amount of the cellular component of the cellular component from the perturbation feature are both downregulated, replacing the significant score with a first score; if the amount of the cellular component of the corresponding cellular component from the single-cell transition feature is downregulated and the amount of the cellular component of the cellular component from the perturbation feature is upregulated, replacing the significant score with a second score; and if the amount of the cellular component of the cellular component from the perturbation feature is not significantly upregulated or downregulated, replacing the significant score with a third score.

[0432] Embodiment 13. The method of any one of Embodiments 1 to 12, wherein the plurality of cellular components comprises a plurality of genes, optionally measured at the RNA level.

[0433] Embodiment 14. A method as described in embodiment 2, wherein each single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets and the second plurality of second single-cell cell component expression datasets is generated using a method selected from the group consisting of: single-cell ribonucleic acid (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq and any combination thereof, and summaries thereof, including combinations representing activated pathways in the single-cell cell component expression datasets, such as linear combinations.

[0434] Embodiment 15. A method as described in any one of Embodiments 1 to 14, wherein the method further comprises: identifying the perturbation as a perturbation that promotes the changed cell state based on the comparison, or identifying the perturbation as a perturbation that inhibits the changed cell state based on the comparison.

[0435] Embodiment 16. A method as described in any one of Embodiments 1 to 15, wherein the cell transformation signature and the perturbation signature are generated using different types of cellular components.

[0436] Embodiment 17. A method as described in any one of embodiments 1 to 16, wherein the cell transformation signature and the perturbation signature are generated using the same type of cellular components.

[0437] Embodiment 18. A method as described in any of Embodiments 1 to 17, wherein the accessing in electronic form is performed for each corresponding feature in a plurality of perturbations to obtain a plurality of perturbation features, and the comparing compares the single-cell transition feature and the perturbation feature with each corresponding feature in a plurality of perturbation features to determine a subset of the plurality of perturbations associated with the transition of the cell between the first cell state and the altered cell state.

[0438] Embodiment 19. A computer system comprising one or more processors and a memory storing instructions for executing the method as described in any one of embodiments 1 to 18.

[0439] Embodiment 20. A non-transitory computer-readable medium storing one or more computer programs executable by a computer for predicting whether a perturbation will affect cell transformation, the computer comprising one or more processors and a memory, the one or more computer programs collectively encoding computer executable instructions for performing the method as described in any one of embodiments 1 to 18.

[0440] Cited References and Alternative Embodiments

[0441] All references cited herein are incorporated by reference in their entirety and for all purposes to the same extent as if each individual publication or patent or patent application was specifically and individually indicated to be incorporated by reference in its entirety for all purposes.

[0442] The present invention may be implemented as a computer program product, which includes a computer program mechanism embedded in a non-transitory computer-readable storage medium. For example, the computer program product may include Figure 1 or Figure 2 The program modules shown in any combination of the above may be stored on a CD-ROM, a DVD, a magnetic disk storage product or any other non-transitory computer-readable data or program storage product.

[0443] As will be apparent to those skilled in the art, many modifications and variations of the present invention may be made without departing from its spirit and scope. The specific embodiments described herein are provided by way of example only. The embodiments have been selected and described to best explain the principles of the present invention and its practical application, so that others skilled in the art can best utilize the present invention and various embodiments with various modifications suitable for the intended specific use. The present invention is limited only by the terms of the appended claims and the full scope of equivalents mentioned in such claims.

[0444] Supplementary Table 1

[0445]

[0446]

[0447]

[0448]

Claims

1. A method for predicting whether a perturbation will affect cell transformation, the method comprising: On a computer system comprising a memory and one or more processors: Accessing a single cell transition signature in electronic form, the single cell transition signature representing a measure of differential cellular component expression between a first cell state and an altered cell state, wherein the altered cell state occurs by the cellular transition from the first cell state to the altered cell state, and wherein the single cell transition signature comprises identification of a plurality of cellular components and, for each respective cellular component in the plurality of cellular components, a corresponding first significance score quantifying an association between (i) a change in expression of the respective cellular component and (ii) a change in cellular state between the first cell state and the altered cell state, and wherein accessing the single cell transition signature comprises determining the single cell transition signature by determining a difference in cellular component amounts of a plurality of cellular components between (i) a first plurality of first single cell cellular component expression datasets and (ii) a second plurality of second single cell cellular component expression datasets using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein obtaining each corresponding first single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets from a corresponding single cell of the first plurality of cells in the first cell state, obtaining each respective second single cell cellular component expression dataset in the second plurality of second single cell cellular component expression datasets from a corresponding single cell of the second plurality of cells in the altered cellular state, and each of the first plurality of first single-cell cellular component expression datasets and the second plurality of second single-cell cellular component expression datasets is generated using a method selected from the group consisting of: single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination thereof; accessing in electronic form a perturbation signature representing a measure of differential cellular component expression between a plurality of unperturbed cells and a plurality of perturbed cells exposed to the perturbation, wherein the perturbation signature comprises an identification of all or a portion of the plurality of cellular components and, for each respective cellular component in the all or the portion of the plurality of cellular components, a corresponding second significance score quantifying an association between (i) a change in expression of the respective cellular component between the plurality of unperturbed cells and the plurality of perturbed cells and (ii) a change in cellular state between the plurality of unperturbed cells and the plurality of perturbed cells, wherein the measure of differential cellular component expression quantifies a difference in the amount of the cellular component between (i) a third plurality of third single cell cellular component expression datasets and (ii) a fourth plurality of fourth single cell cellular component expression datasets using one of a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein: obtaining each respective third single cell cellular component expression dataset in the third plurality of third single cell cellular component expression datasets from a corresponding single cell in the plurality of unperturbed cells, and obtaining each respective fourth single cell cellular component expression dataset in the fourth plurality of fourth single cell cellular component expression datasets from a corresponding single cell of a fourth plurality of cells in the plurality of perturbed cells exposed to the perturbation; and A first significance score in the single cell transition signature is compared to a corresponding second significance score in the perturbation signature to determine whether the perturbation will affect the cell transition.

2. The method of claim 1, wherein: Each respective dataset of the first plurality of single-cell cellular component expression datasets comprises a corresponding cellular component vector of the first plurality of cellular component vectors, Each respective dataset of the second plurality of single-cell cellular component expression datasets comprises a corresponding cellular component vector of the second plurality of cellular component vectors, Each respective cellular component vector in the first plurality of cellular component vectors and the second plurality of cellular component vectors comprises a plurality of elements, each respective element in the respective cellular component vectors being associated with a respective cellular component in the plurality of cellular components and comprising a respective value representing the amount of the respective cellular component in the respective single cell in the first plurality of cells or the second plurality of cells.

3. The method of claim 2, further comprising: performing dimensionality reduction on the first plurality of single-cell cellular component expression datasets and / or the second plurality of single-cell cellular component expression datasets to generate a plurality of reduced-dimensionality components; for each respective cellular component vector in the first plurality of cellular component vectors and the second plurality of cellular component vectors, applying the plurality of reduced dimensionality components to the respective cellular component vector to form a corresponding reduced dimensionality vector, the corresponding reduced dimensionality vector comprising a reduced dimensionality component value of each respective reduced dimensionality component in the plurality of reduced dimensionality components, thereby forming a corresponding first plurality of reduced dimensionality vectors and a second plurality of reduced dimensionality vectors; as well as Perform clustering to generate clusters C j a set of clusters, each cluster comprising a plurality of points corresponding to a subset of the first plurality of reduced dimensionality vectors and the second plurality of reduced dimensionality vectors; From the cluster C j a first cluster of the set identifying the first plurality of cells; as well as From the cluster C j The second cluster of the set identifies the second plurality of cells, The method optionally further comprises performing manifold learning with the corresponding first plurality of reduced dimensionality vectors and second plurality of reduced dimensionality vectors to identify a relative cellular state of each cell relative to each other cell in the first plurality of cells and the second plurality of cells.

4. The method of claim 1, wherein the plurality of unperturbed cells are control cells that have not been exposed to the perturbation, or wherein the unperturbed cells are an average of unrelated perturbed cells that have been exposed to the perturbation.

5. The method of claim 1, further comprising: The single cell transition signature and the perturbation signature are pruned to limit the plurality of cellular components to transcription factors, optionally measured at the RNA level.

6. The method of claim 1, further comprising: filtering the single cell transition feature and the perturbation feature to reduce the number of cellular components included in the single cell transition feature and the perturbation feature, Optionally, the filtering of the single cell transition signature and the perturbation signature comprises reducing the number of cellular components included in the single cell transition signature and the perturbation signature according to a threshold p value or according to a threshold number of cellular components.

7. The method of claim 1, wherein determining the corresponding second significance score for the respective cellular component comprises: For each corresponding cellular component in the plurality of cellular components, replacing the significance score of the corresponding cellular component with the corresponding matching score of the corresponding cellular component; combining the match scores for the plurality of cellular components to generate a match score for the perturbation; as well as determining whether the corresponding perturbation is associated with the transition of a cell between the first cell state and the altered cell state based on the matching score of the perturbation, Optionally wherein the corresponding match scores comprise discrete or continuous scores.

8. The method of claim 7, wherein replacing the significance score comprises: if the amount of the cellular component of the corresponding cellular component from the single-cell transition feature and the amount of the cellular component of the corresponding cellular component from the perturbation feature are both upregulated, replacing the significance score with the first score; replacing the significance score with a second score if the amount of the cellular component from the single-cell transition signature of the corresponding cellular component is upregulated and the amount of the cellular component from the perturbation signature of the corresponding cellular component is downregulated; as well as If the amount of the cellular component from the perturbation feature of the corresponding cellular component is not significantly upregulated or downregulated, replacing the significant score with a third score.

9. The method of claim 7, wherein replacing the saliency score comprises: if both the amount of the cellular component from the single-cell transition feature and the amount of the cellular component from the perturbation feature of the corresponding cellular component are downregulated, replacing the significance score with the first score; replacing the significance score with a second score if the amount of the cellular component from the single-cell transition signature of the corresponding cellular component is downregulated and the amount of the cellular component from the perturbation signature of the cellular component is upregulated; as well as If the amount of the cellular component from the perturbation feature of the cellular component is not significantly upregulated or downregulated, replacing the significant score with a third score.

10. The method of claim 1, wherein the plurality of cellular components comprises a plurality of genes, optionally measured at the RNA level.

11. The method of claim 1, further comprising: identifying the perturbation as one that promotes the altered cellular state based on the comparison, or The perturbation is identified based on the comparison as a perturbation that inhibits the altered cellular state.

12. The method of claim 1, wherein the cell transition signature and the perturbation signature are generated using different types of cellular components.

13. The method of claim 1, wherein the cell transition signature and the perturbation signature are generated using the same type of cellular components.

14. The method of claim 1, wherein performing said accessing in electronic form for each corresponding characteristic in a plurality of perturbations, thereby obtaining a plurality of perturbation characteristics, and The comparing compares the single-cell transition signature and the perturbation signature to each corresponding signature in a plurality of perturbation signatures to thereby determine a subset of the plurality of perturbations associated with a transition of a cell between the first cell state and the altered cell state.

15. A computer system comprising one or more processors and a memory storing instructions for executing a method for predicting whether a perturbation will affect a cell transition, the method comprising: Accessing a single cell transition signature in electronic form, the single cell transition signature representing a measure of differential cellular component expression between a first cell state and an altered cell state, wherein the altered cell state occurs by the cellular transition from the first cell state to the altered cell state, and wherein the single cell transition signature comprises identification of a plurality of cellular components and, for each respective cellular component in the plurality of cellular components, a corresponding first significance score quantifying an association between (i) a change in expression of the respective cellular component and (ii) a change in cellular state between the first cell state and the altered cell state, and wherein accessing the single cell transition signature comprises determining the single cell transition signature by determining a difference in cellular component amounts of a plurality of cellular components between (i) a first plurality of first single cell cellular component expression datasets and (ii) a second plurality of second single cell cellular component expression datasets using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein obtaining each corresponding first single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets from a corresponding single cell of the first plurality of cells in the first cell state, obtaining each respective second single cell cell component expression dataset in the second plurality of second single cell cell component expression datasets from a corresponding single cell of the second plurality of cells in the altered cell state, and each of the first plurality of first single-cell cellular component expression datasets and the second plurality of second single-cell cellular component expression datasets is generated using a method selected from the group consisting of: single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination thereof; accessing in electronic form a perturbation signature representing a measure of differential cellular component expression between a plurality of unperturbed cells and a plurality of perturbed cells exposed to the perturbation, wherein the perturbation signature comprises an identification of all or a portion of the plurality of cellular components and, for each respective cellular component in the all or the portion of the plurality of cellular components, a corresponding second significance score quantifying an association between (i) a change in expression of the respective cellular component between the plurality of unperturbed cells and the plurality of perturbed cells and (ii) a change in cellular state between the plurality of unperturbed cells and the plurality of perturbed cells, wherein the measure of differential cellular component expression quantifies a difference in the amount of the cellular component between (i) a third plurality of third single cell cellular component expression datasets and (ii) a fourth plurality of fourth single cell cellular component expression datasets using one of a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein: obtaining each respective third single cell cellular component expression dataset in the third plurality of third single cell cellular component expression datasets from a corresponding single cell in the plurality of unperturbed cells, and obtaining each respective fourth single cell cellular component expression dataset in the fourth plurality of fourth single cell cellular component expression datasets from a corresponding single cell of a fourth plurality of cells in the plurality of perturbed cells exposed to the perturbation; and A first significance score in the single cell transition signature is compared to a corresponding second significance score in the perturbation signature to determine whether the perturbation will affect the cell transition.

16. A non-transitory computer readable medium storing one or more computer programs executable by a computer for predicting whether a perturbation will affect a cell transition, the computer comprising one or more processors and memory, the one or more computer programs collectively encoding computer executable instructions for performing the following method, the method comprising: Accessing a single cell transition signature in electronic form, the single cell transition signature representing a measure of differential cellular component expression between a first cell state and an altered cell state, wherein the altered cell state occurs by the cellular transition from the first cell state to the altered cell state, and wherein the single cell transition signature comprises identification of a plurality of cellular components and, for each respective cellular component in the plurality of cellular components, a corresponding first significance score quantifying an association between (i) a change in expression of the respective cellular component and (ii) a change in cellular state between the first cell state and the altered cell state, and wherein accessing the single cell transition signature comprises determining the single cell transition signature by determining a difference in cellular component amounts of a plurality of cellular components between (i) a first plurality of first single cell cellular component expression datasets and (ii) a second plurality of second single cell cellular component expression datasets using one of a mean difference test, a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein obtaining each corresponding first single-cell cell component expression dataset in the first plurality of first single-cell cell component expression datasets from a corresponding single cell of the first plurality of cells in the first cell state, obtaining each respective second single cell cellular component expression dataset in the second plurality of second single cell cellular component expression datasets from a corresponding single cell of the second plurality of cells in the altered cellular state, and each of the first plurality of first single-cell cellular component expression datasets and the second plurality of second single-cell cellular component expression datasets is generated using a method selected from the group consisting of: single-cell RNA (RNA) sequencing (scRNA-seq), scTag-seq, single-cell assay of transposase-accessible chromatin using sequencing (scATAC-seq), CyTOF / SCoP, E-MS / Abseq, miRNA-seq, CITE-seq, and any combination thereof; accessing in electronic form a perturbation signature representing a measure of differential cellular component expression between a plurality of unperturbed cells and a plurality of perturbed cells exposed to the perturbation, wherein the perturbation signature comprises an identification of all or a portion of the plurality of cellular components and, for each respective cellular component in the all or the portion of the plurality of cellular components, a corresponding second significance score quantifying an association between (i) a change in expression of the respective cellular component between the plurality of unperturbed cells and the plurality of perturbed cells and (ii) a change in cellular state between the plurality of unperturbed cells and the plurality of perturbed cells, wherein the measure of differential cellular component expression quantifies a difference in the amount of the cellular component between (i) a third plurality of third single cell cellular component expression datasets and (ii) a fourth plurality of fourth single cell cellular component expression datasets using one of a Wilcoxon rank sum test, a t-test, a logistic regression, and a generalized linear model, wherein: obtaining each respective third single cell cellular component expression dataset in the third plurality of third single cell cellular component expression datasets from a corresponding single cell in the plurality of unperturbed cells, and obtaining each respective fourth single cell cellular component expression dataset in the fourth plurality of fourth single cell cellular component expression datasets from a corresponding single cell of a fourth plurality of cells in the plurality of perturbed cells exposed to the perturbation; and A first significance score in the single cell transition signature is compared to a corresponding second significance score in the perturbation signature to determine whether the perturbation will affect the cell transition.

17. The method of claim 1, wherein each corresponding second significance score is a p-value or a t-value.

Images