Identifying disease subtypes

Abstract

The systems and methods provided herein relate to identifying disease subtypes based on gene expression data. For example, the methods can include defining, based on gene expression data for a plurality of target genes, one or more gene expression signatures corresponding to one or more endotypes of the disease; measuring expression levels of each of a plurality of genes in a biological sample from the subject; generating, based on expression levels of each of the plurality of genes in the biological sample from the subject, a gene expression signature for the subject; comparing the gene expression signature for the subject to the one or more gene expression signatures corresponding to one or more endotypes of the disease; and assigning, based on the comparison of gene expression signatures, a disease endotype to the subject having the disease.

Description

[0001] IDENTIFYING DISEASE SUBTYPES

[0002] CLAIM OF PRIORITY

[0003] This application claims priority to European Application No. 23315401.2, filed on October 27, 2023. The entire contents of the foregoing are incorporated herein by reference.

[0004] TECHNICAL FIELD

[0005] This disclosure relates to identifying disease subtypes based on gene expression data.

[0006] BACKGROUND

[0007] Complex diseases can exhibit a significant degree of heterogeneity, which is reflected in the presence of patient subpopulations that display distinct pathophysiology, disease progression, and response to therapeutic treatments. This heterogeneity represents a significant barrier for translational efforts that aim to identify novel therapies and to design and conduct clinical trials. A deeper understanding of diseases in terms of endotypes (underlying cellular and molecular mechanisms) as opposed to phenotypes (observable clinical / pathological characteristics) has two important impacts in drug discovery. First, it accelerates the implementation of more effective and targeted therapies, directed to specific disease endotypes with shared molecular mechanisms. Second, it enables the pre-selection of patients that, according to their disease endotype, are best poised to respond to a given therapy in clinical trials, therefore increasing the probability of success of the trial. Disease endotypes can be interrogated by, for example, high-throughput gene expression data, including RNA-seq data. There is therefore a need for systems and methods for efficient identification of disease endotypes based on high- throughput gene expression data.

[0008] SUMMARY

[0009] Provided herein are systems and methods for identifying disease endotypes based on high-throughput gene expression data, e.g., transcriptome data. The systems and methods provided herein can be applied to diseases with relevant tissue transcriptome profiling data. Reproducible disease endotypes can be identified with this approach and relevant disease endotypes (or patient stratification) can be included as a part of a precision medicine strategy to assess the drug treatment responses based on disease endotypes. The systems and methods provided herein can include a computational framework for unbiased transcriptomic data-driven disease subtype identification, reproducibility assessment and phenotypic characterization. The inclusion of multiple cohorts and systematic matching to identify reproducible subtypes make this approach very rigorous, and results are biologically interpretable. The systems and methods provided herein can include an unsupervised machine learning approach for processing gene expression data to identify disease endotypes to support precision medicine strategies and development of targeted therapeutics. The systems and methods provide an experimental and computational analytical framework aimed at utilizing drug signature genes for i) the identification of reproducible disease endotypes; ii) the prediction of response to specific drugs in each patient endotype; and iii) the discovery of a genomicsbased classifier to assign patients to endotypes, for translation to stratified clinical trials.

[0010] In a first aspect, the disclosure provides methods of determining a disease endotype for a subject having the disease, the method including defining, based on gene expression data for a plurality of target genes, one or more gene expression signatures corresponding to one or more endotypes of the disease; measuring expression levels of each of a plurality of genes in a biological sample from the subject; generating, based on expression levels of each of the plurality of genes in the biological sample from the subject, a gene expression signature for the subject; comparing the gene expression signature for the subject to the one or more gene expression signatures corresponding to one or more endotypes of the disease; and assigning, based on the comparison of gene expression signatures, a disease endotype to the subject having the disease. In some embodiments, the methods further include predicting a response to an exogenous treatment based on the disease endotype assigned to the subject having the disease. In some embodiments, the gene expression data is RNA-seq data. In some embodiments, the gene expression data for a plurality of target genes is from two or more previously generated datasets. In some embodiments, the defining step further includes use of a machine learning algorithm. In some embodiments, the machine learning algorithm further includes K-means clustering. In some embodiments, the gene expression signature utilizes K-means clustering. In some embodiments, the disease is a cancer. In some embodiments, the disease is an autoimmune disease. In some embodiments, the biological sample is a blood sample. In some embodiments, the biological sample is a urine sample.

[0011] In another aspect, the disclosure provides methods of treating a disease in a subject, the methods including obtaining an expression level of each of a plurality of genes in a biological sample obtained from the subject; assigning the disease of the subject to a disease endotype based on the expression level of each of the plurality of genes in the biological sample, the disease endotype being selected from a set of endotypes previously determined from gene expression data derived from multiple datasets; administering a therapeutic agent to the subject, wherein the assigning the disease of the subject to the disease endotype indicates that the therapeutic agent is predicted to be effective for treating the disease in the subject. In some embodiments, the biological sample is a blood sample. In some embodiments, the biological sample is a urine sample. In some embodiments, the level of expression of one or more of the plurality of genes of the disease endotype is increased relative to genes that are not assigned to the disease endotype. In some embodiments, the level of expression of one or more of the plurality of genes of the disease endotype is decreased relative to genes that are not assigned to the disease endotype.

[0012] In another aspect, the disclosure provides computer-implemented methods for determining a disease endotype for a subject having the disease, the method including receiving, by a computing device including a processor programmed to execute software instructions within a memory, a set of gene expression data comprising expression levels of each of a plurality of genes in a biological sample obtained from the subject; applying a machine learning algorithm to generate a classification model used in a classification scheme, wherein the machine learning algorithm has been trained using a stratified traintest split of the set of gene expression data; applying, by the computing device, the classification scheme, to rank the set of gene expression data; generating, by the computing device, a reduced set of genes that distinguish disease endotypes; determining, based on the expression levels of the reduced set of genes that distinguish disease endotypes, the disease endotype for the subject having the disease. In some embodiments, the methods further include predicting a response to an exogenous treatment based on the disease endotype determined for the subject having the disease. In some embodiments, the machine learning algorithm comprises feature selection. In some embodiments, the feature selection is forward feature group selection (FFGS).

[0013] In another aspect, the disclosure provides a computing system for determining a disease endotype for a subject having the disease, the computing system including: a data server for receiving gene expression data for a plurality of target genes and measurements of expression levels of each of a plurality of genes in a biological sample from the subject; a computing device communicatively connected to the data server, the computing device comprising an application server configured to: define, based on the gene expression data, one or more gene expression signatures corresponding to one or more endotypes of the disease; generate, based on the expression levels of each of the plurality of genes in the biological sample from the subject, a gene expression signature for the subject; compare the gene expression signature for the subject to the one or more gene expression signatures corresponding to one or more endotypes of the disease; and assign, based on the comparison of gene expression signatures, a disease endotype to the subject having the disease; and a display communicatively connected to the computing device and configured to display a report describing the assignment of the disease endotype to the subject having the disease.

[0014] In some embodiments, the application server is further configured to predict a response to an exogenous treatment based on the disease endotype assigned to the subject having the disease. In some embodiments, the gene expression data is RNA-seq data. In some embodiments, the gene expression data for a plurality of target genes is from two or more previously generated datasets. In some embodiments, the defining step further includes use of a machine learning algorithm. In some embodiments, the machine learning algorithm further includes K-means clustering. In some embodiments, generating the gene expression signature utilizes K-means clustering. In some embodiments, the disease is a cancer. In some embodiments, the disease is an autoimmune disease. In some embodiments, the biological sample is a blood sample. In some embodiments, the biological sample is a urine sample.

[0015] Unless otherwise defined, all technical and scientific terms used herein have the same meaning as commonly understood by one of ordinary skill in the art to which this invention belongs. Methods and materials are described herein for use in the present invention; other, suitable methods and materials known in the art can also be used. The materials, methods, and examples are illustrative only and not intended to be limiting. All publications, patent applications, patents, sequences, database entries, and other references mentioned herein are incorporated by reference in their entirety. In case of conflict, the present specification, including definitions, will control.

[0016] Other features and advantages of the invention will be apparent from the following detailed description and figures, and from the claims.

[0017] DESCRIPTION OF DRAWINGS

[0018] The patent or application file contains at least one drawing executed in color. Copies of this patent or patent application publication with color drawing(s) will be provided by the Office upon request and payment of the necessary fee.

[0019] FIG. 1A is a schematic of step of the methods disclosed herein, wherein a biological sample is obtained from a subject and treated with an activating agent, e.g., a potential therapeutic. The biological sample can comprise multiple cell types, and gene expression signatures can be produced from gene expression data generated from the biological sample before or after treating the biological sample with the activating agent.

[0020] FIG. IB is a schematic of a step of the methods disclosed herein, wherein gene expression signatures of subgroups of patients are derived from multiple datasets of gene expression data. Specifically, clustering analysis is performed and clusters are systematically matched and validated across datasets to produce multiple disease endotypes based on the gene expression signatures. FIG. 2A is a schematic of a step of the methods disclosed herein, wherein based on the gene expression signatures and associated disease endotypes, patients can be assigned to disease endotypes based on the gene expression data of the individual patient.

[0021] FIG. 2B is a schematic of a step of the methods disclosed herein, wherein a biological sample is obtained from a subject, gene expression data are generated from the biological sample, and based on the gene expression signatures and associated disease endotypes, a therapeutic agent predicted to be effective for a specific disease endotype can be administered to a patient with that disease endotype.

[0022] FIG. 3A is a schematic of a step of the methods disclosed herein, wherein disease endotypes are determined from gene expression data derived from each dataset.

[0023] FIG. 3B is a schematic of a step of the methods disclosed herein, wherein systematic disease endotype matching is performed across the multiple datasets.

[0024] FIG. 3C is a schematic of a step of the methods disclosed herein, wherein each disease endotype is characterized by its gene expression signature.

[0025] FIG. 4A is a schematic of a step of the methods disclosed herein, wherein multiple disease endotypes are determined from gene expression data derived from multiple datasets, and a subset of disease endotypes are validated across the multiple datasets to confirm the validity of the disease endotypes.

[0026] FIG. 4B is a schematic of a step of the methods disclosed herein, wherein gene expression signatures of cell types are compared across the multiple datasets in order to characterize and validate the determined disease endotypes.

[0027] FIG. 5 is a diagram of computer system components that can be used to implement methods for identifying disease subtypes based on gene expression data.

[0028] FIG. 6 is a diagram of building, applying, and assessing the performance of a classifier using a study cohort as a training dataset and a testing dataset.

[0029] DETAILED DESCRIPTION

[0030] Determining Gene Expression Signatures for Disease Endotypes

[0031] The present disclosure provides systems and methods for identifying disease endotypes based on high-throughput gene expression data, e.g., transcriptome data comprising expression-based analysis of a plurality of targets. Generally, the method comprises (a) developing an activation signature based on response to an exogenous treatment; (b) determining reproducible disease endotypes from multiple gene expression datasets and expression levels of a plurality of targets; (c) evaluating the activation signature across multiple disease endotypes; and (d) assigning the exogenous treatment to one or more disease endotypes based on activation scores of signatures of the disease endotypes. Assaying the expression level for a plurality of targets in the sample may comprise applying the sample to a microarray. Assaying the expression level for a plurality of targets in the sample may comprise generating RNA-seq data. In some instances, assaying the expression level may comprise the use of an algorithm. The algorithm may be used to produce a classifier. Alternatively, the classifier may provide a probe selection region. In some instances, assaying the expression level for a plurality of targets comprises detecting and / or quantifying the plurality of targets. In some embodiments, assaying the expression level for a plurality of targets comprises sequencing and quantifying the plurality of targets. In some embodiments, assaying the expression level for a plurality of targets comprises amplifying the plurality of targets. In some embodiments, assaying the expression level for a plurality of targets comprises quantifying the plurality of targets. In some embodiments, assaying the expression level for a plurality of targets comprises conducting a multiplexed reaction on the plurality of targets.

[0032] In some embodiments, as illustrated in the schematic of FIG. 1A, a biological sample can be obtained from a subject and treated with an activating agent, e.g., a potential therapeutic. The biological sample can comprise multiple cell types (e.g., cell types A, B, C, D, and E in FIG. 1A), and gene expression signatures can be produced from gene expression data generated from the biological sample before or after treating the biological sample with the activating agent. As illustrated in FIG. IB, a gene expression dataset can be generated for each cell-type, and after cluster matching and validation, reproducible disease subtypes can be generated for each cell-type, based on the response to the activating agent.

[0033] As illustrated in FIG. 2A, based on the gene expression signatures and associated disease endotypes, patients can be assigned to disease endotypes based on the gene expression data of the individual patient. As illustrated in FIG. 2B, a biological sample can be obtained from a subject, gene expression data can be generated from the biological sample, and based on the gene expression signatures and associated disease endotypes, a therapeutic agent predicted to be effective for a specific disease endotype can be administered to a patient with that disease endotype.

[0034] In some embodiments, as illustrated in the schematic of FIG. 3A, disease endotypes are determined from gene expression data derived from each dataset in an unbiased way. In some embodiments, disease endotypes are determined from 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more datasets (see, FIG. 3A). In some embodiments, systematic disease endotype matching is performed across the multiple datasets (see, FIG. 3B). In some embodiments, each disease endotype is characterized by its gene expression signature (see, FIG. 3C)

[0035] In some embodiments, the high-throughput gene expression data, e.g., transcriptome data comprising expression-based analysis of a plurality of targets is generated from a biological sample. In some embodiments, the biological sample comprises nucleic acids (e.g., RNA or DNA). In some instances, the plurality of targets comprises at least about 2, at least about 3, at least about 4, at least about 5, at least about 6, at least about 7, at least about 8, at least about 9, or at least about 10 targets. In some instances, the plurality of targets comprises a coding target, non-coding target, or any combination thereof. In some instances, the coding target comprises an exonic sequence. In other instances, the non-coding target comprises a non-exonic or exonic sequence. Alternatively, a non-coding target comprises a UTR sequence, an intronic sequence, antisense, or a non-coding RNA transcript. In some instances, a non-coding target comprises sequences which partially overlap with a UTR sequence or an intronic sequence. A non-coding target also includes non-exonic and / or exonic transcripts. Exonic sequences may comprise regions on a protein-coding gene, such as an exon, UTR, or a portion thereof. Non-exonic sequences may comprise regions on a protein-coding, non- protein-coding gene, or a portion thereof. For example, non-exonic sequences may comprise intronic regions, promoter regions, intergenic regions, a non-coding transcript, an exon anti-sense region, an intronic anti-sense region, UTR anti-sense region, noncoding transcript anti-sense region, or a portion thereof. In other instances, the plurality of targets comprises a noncoding RNA transcript.

[0036] The plurality of targets may comprise one or more targets selected from a classifier disclosed herein. The classifier may be generated from one or more models or algorithms. The one or more models or algorithms may be Naive Bayes (NB), AdaBoost (Adaptive Boosting), recursive Partitioning (Rpart), random forest (RF), support vector machine (SVM), k-nearest neighbor (KNN), high dimensional discriminate analysis (HDD A), or a combination thereof. The classifier may have an area under the receiver operating characteristic (ROC) curve (AUC) of equal to or greater than 0.60. The classifier may have an AUC of equal to or greater than 0.61. The classifier may have an AUC of equal to or greater than 0.62. The classifier may have an AUC of equal to or greater than 0.63. The classifier may have an AUC of equal to or greater than 0.64. The classifier may have an AUC of equal to or greater than 0.65. The classifier may have an AUC of equal to or greater than 0.66. The classifier may have an AUC of equal to or greater than 0.67. The classifier may have an AUC of equal to or greater than 0.68. The classifier may have an AUC of equal to or greater than 0.69. The classifier may have an AUC of equal to or greater than 0.70. The classifier may have an AUC of equal to or greater than 0.75. The classifier may have an AUC of equal to or greater than 0.77. The classifier may have an AUC of equal to or greater than 0.78. The classifier may have an AUC of equal to or greater than 0.79. The classifier may have an AUC of equal to or greater than 0.80. The AUC may be clinically significant based on its 95% confidence interval (CI). The accuracy of the classifier may be at least about 70%. The accuracy of the classifier may be at least about 73%. The accuracy of the classifier may be at least about 75%. The accuracy of the classifier may be at least about 77%. The accuracy of the classifier may be at least about 80%. The accuracy of the classifier may be at least about 83%. The accuracy of the classifier may be at least about 84%. The accuracy of the classifier may be at least about 86%. The accuracy of the classifier may be at least about 88%. The accuracy of the classifier may be at least about 90%. The p-value of the classifier may be less than or equal to 0.05. The p-value of the classifier may be less than or equal to 0.04. The p-value of the classifier may be less than or equal to 0.03. The p- value of the classifier may be less than or equal to 0.02. The p-value of the classifier may be less than or equal to 0.01 . The p-value of the classifier may be less than or equal to 0.008. The p-value of the classifier may be less than or equal to 0.006. The p-value of the classifier may be less than or equal to 0.004. The p-value of the classifier may be less than or equal to 0.002. The p-value of the classifier may be less than or equal to 0.001.

[0037] The plurality of targets may comprise one or more targets selected from a Random Forest (RF) classifier. The plurality of targets may comprise two or more targets selected from a Random Forest (RF) classifier. The plurality of targets may comprise three or more targets selected from a Random Forest (RF) classifier. The plurality of targets may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 20, 25, 30, 35, 40, 45, 50, or more targets selected from a Random Forest (RF) classifier. The RF classifier may be an RF2, and RF3, or an RF4 classifier. The RF classifier may be an RF22 classifier (e g., a Random Forest classifier with 22 targets).

[0038] The plurality of targets may comprise one or more targets selected from an SVM classifier. The plurality of targets may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10 or more targets selected from an SVM classifier. The plurality of targets may comprise 12, 13, 14, 15, 17, 20, 22, 25, 27, 30 or more targets selected from an SVM classifier. The plurality of targets may comprise 32, 35, 37, 40, 43, 45, 47, 50, 53, 55, 57, 60 or more targets selected from an SVM classifier. The SVM classifier may be an SVM2 classifier.

[0039] The plurality of targets may comprise one or more targets selected from a KNN classifier. The plurality of targets may comprise 2, 3, 4, 5, 6, 7, 8, 9, 10 or more targets selected from a KNN classifier. The plurality of targets may comprise 12, 13, 14, 15, 17, 20, 22, 25, 27, 30 or more targets selected from a KNN classifier. The plurality of targets may comprise 32, 35, 37, 40, 43, 45, 47, 50, 53, 55, 57, 60 or more targets selected from a KNN classifier. The plurality of targets may comprise 65, 70, 75, 80, 85, 90, 95, 100 or more targets selected from a KNN classifier.

[0040] In some embodiments, one or more pattern recognition methods can be used in analyzing the expression level of target sequences. The pattern recognition method can comprise a linear combination of expression levels, or a nonlinear combination of expression levels. In some embodiments, expression measurements for RNA transcripts or combinations of RNA transcript levels are formulated into linear or non-linear models or algorithms (e.g., an 'expression signature') and converted into a likelihood score. This likelihood score may indicate the probability that a biological sample is from a patient who will benefit from a disease endotype-specific therapy.

[0041] Additionally, a likelihood score may indicate the probability that a biological sample is from a patient who may exhibit endotype-specific prognosis or response to treatment. The likelihood score can be used to distinguish these disease states. The models and / or algorithms can be provided in machine readable format, and may be used to correlate expression levels or an expression profile with a disease state, and / or to designate a treatment modality for a patient or class of patients.

[0042] Assaying the expression level for a plurality of targets may comprise the use of an algorithm or classifier. High-throughput gene expression data, e.g. array data or RNA-seq data, can be managed, classified, and analyzed using techniques known in the art. Assaying the expression level for a plurality of targets may comprise probe set modeling and data pre-processing. Probe set modeling and data pre-processing can be derived using the Robust Multi-Array (RMA) algorithm or variants GC-RMA, fRMA, Probe Logarithmic Intensity Error (PLIER) algorithm, or variant iterPLIER, or Single-Channel Array Normalization (SCAN) algorithm.

[0043] Variance or intensity filters can be applied to pre-process data using the RMA algorithm, for example by removing target sequences with a standard deviation of <10 or a mean intensity of <1 00 intensity units of a normalized data range, respectively.

[0044] Alternatively, assaying the expression level for a plurality of targets may comprise the use of a machine learning algorithm. The machine learning algorithm may comprise a supervised learning algorithm. Examples of supervised learning algorithms may include Average One-Dependence Estimators (AODE), Artificial neural network (e g., Backpropagation), Bayesian statistics (e.g., Naive Bayes classifier, Bayesian network, Bayesian knowledge base), Casebased reasoning, Decision trees, Inductive logic programming, Gaussian process regression, Group method of data handling (GMDH), Learning Automata, Learning Vector Quantization, Minimum message length (decision trees, decision graphs, etc.), Lazy learning, Instance-based learning Nearest Neighbor Algorithm, Analogical modeling, Probably approximately correct learning (PAC) learning, Ripple down rules, a knowledge acquisition methodology, Symbolic machine learning algorithms, Subsymbolic machine learning algorithms, Support vector machines, Random Forests, Ensembles of classifiers, Bootstrap aggregating (bagging), and Boosting. Supervised learning may comprise ordinal classification such as regression analysis and Information fuzzy networks (IFN). Alternatively, supervised learning methods may comprise statistical classification, such as AODE, Linear classifiers (e.g., Fisher's linear discriminant, Logistic regression, Naive Bayes classifier, Perceptron, and Support vector machine), quadratic classifiers, k-nearest neighbor, Boosting, Decision trees (e.g., C4.5, Random forests), Bayesian networks, and Hidden Markov models.

[0045] The machine learning algorithms may also comprise an unsupervised learning algorithm. Examples of unsupervised learning algorithms may include Artificial Neural Network, Data clustering, Expectation-maximization algorithm, Self-organizing map, Radial basis function network, Vector Quantization, Generative topographic map, Information bottleneck method, and IBSEAD. Unsupervised learning may also comprise association rule learning algorithms such as Apriori algorithm, Eclat algorithm and PP- growth algorithm. Hierarchical clustering, such as Single-linkage clustering and Conceptual clustering, may also be used. Alternatively, unsupervised learning may comprise partitional clustering such as K-means algorithm and Fuzzy clustering. In some instances, the machine learning algorithms comprise a reinforcement learning algorithm. Examples of reinforcement learning algorithms include, but are not limited to, temporal difference learning, Q-learning and Learning Automata. Alternatively, the machine learning algorithm may comprise Data Pre-processing.

[0046] In some embodiments, the machine learning algorithms may include, but are not limited to, Average One-Dependence Estimators (AODE), Fisher's linear discriminant, Logistic regression, Perceptron, Multilayer Perceptron, Artificial Neural Networks, Support vector machines, Quadratic classifiers, Boosting, Decision trees, C4.5, Bayesian networks, Hidden Markov models, High-Dimensional Discriminant Analysis, and Gaussian Mixture Models. The machine learning algorithm may comprise support vector machines, Naive Bayes classifier, k-nearest neighbor, high-dimensional discriminant analysis, or Gaussian mixture models. In some instances, the machine learning algorithm comprises Random Forests.

[0047] Molecular subtyping is a method of classifying a disease into one of multiple genetically-distinct categories, or endotypes. Each endotype may respond differently to different kinds of treatments, and the presence of a particular endotype can be predictive of, for example, predicted effectiveness of a certain therapeutic, higher risk of recurrence, or good or poor prognosis of the disease for an individual. As described herein, each endotype has a unique molecular and clinical signature and a characteristic gene expression signature. In some instances, the molecular subtyping methods of the systems and methods provided herein are used in combination with other biomarkers for analyzing the disease endotype.

[0048] Disease endotype subtyping can be utilized to predict whether or not a patient can be predicted to benefit from a particular therapy. For example, patients whose cells respond to a particular activating therapeutic ex vivo may respond well to that therapeutic clinically, and a gene expression signature can be detected after an ex vivo experiment and generalized to additional patients having the same disease endotype as assigned by the gene expression signature. Therefore, patients having the gene expression signature corresponding to response to the activating therapeutic ex vivo can be assigned to a disease endotype indicating a particular clinical therapeutic, whereas patients having the gene expression signature corresponding to a lack of response to the activating therapeutic ex vivo may be better candidates for other clinical therapeutic options.

[0049] As used herein a "signature" or "gene signature" may encompass any gene or genes, protein or proteins, or epigenetic element(s) whose expression profile or whose occurrence is associated with a specific cell type, subtype, or cell state of a specific cell type or subtype within a population of cells. For ease of discussion, when discussing gene expression, any of gene or genes, protein or proteins, or epigenetic element(s) may be substituted. As used herein, the terms "signature", "expression profile", or "expression program" may be used interchangeably. It is to be understood that also when referring to proteins (e.g., differentially expressed proteins), such may fall within the definition of "gene" signature. Levels of expression or activity or prevalence may be compared between different cells in order to characterize or identify for instance signatures specific for cell subpopulations. Increased or decreased expression or activity or prevalence of signature genes may be compared between different cells in order to characterize or identify, for instance, specific cell subpopulations. The detection of a signature in single cells may be used to identify and quantitate, for instance, specific cell subpopulations. A signature may include a gene or genes, protein or proteins, or epigenetic element(s) whose expression or occurrence is specific to a cell subpopulation, such that expression or occurrence is exclusive to the cell subpopulation. A gene signature as used herein, may thus refer to any set of up- and down-regulated genes that are representative of a cell type or subtype. A gene signature as used herein, may also refer to any set of up- and down- regulated genes between different cells or cell subpopulations derived from a geneexpression profile. For example, a gene signature may comprise a list of genes differentially expressed in a distinction of interest.

[0050] The detection of a signature can be performed in cells of a patient ex vivo after treatment with a potential therapeutic. The detection of a signature may be used to identify and quantitate, for instance, response of the cells to treatment with the potential therapeutic. A signature may include a gene or genes, protein or proteins, or epigenetic element(s) whose expression or occurrence is specific cells that respond to the potential therapeutic, such that expression or occurrence is exclusive to the cells that respond to the therapeutic. A gene signature as used herein, may thus refer to any set of up- and down- regulated genes that are representative of cells that respond to the potential therapeutic. For example, a gene signature may comprise a list of genes differentially expressed in cells that respond to the potential therapeutic ex vivo relative to cells that do not respond to the potential therapeutic ex vivo.

[0051] The signature as defined herein (being it a gene signature, protein signature or other genetic or epigenetic signature) can be used to indicate the presence of a cell type, a subtype of the cell type, the state of the microenvironment of a population of cells, a particular cell type population or subpopulation, and / or the overall status of the entire cell subpopulation. Furthermore, the signature may be indicative of cells within a population of cells in vivo. The signature may also be used to suggest for instance particular therapies, or to follow up treatment, or to suggest ways to modulate immune systems. The signatures of the present invention may be discovered by analysis of expression profiles of single cells within a population of cells from isolated samples (e.g., cells isolated from a patient), thus allowing the discovery of novel cell subtypes or cell states that were previously invisible or unrecognized. The presence of subtypes or cell states may be determined by subtype specific or cell state specific signatures. The presence of these specific cell subtypes or cell states may be determined by applying the signature genes to bulk sequencing data or array data in a sample. Not being bound by any particular theory, the signatures of the systems and method disclosed herein may be microenvironment specific, such as their expression in a particular spatio-temporal context. Not being bound by any particular theory, signatures as discussed herein are specific to a particular pathological context. Not being bound by any particular theory, a combination of cell subtypes having a particular signature may indicate an outcome. Not being bound by any particular theory, the signatures can be used to deconvolute the network of cells present in a particular pathological condition. Not being bound by any particular theory, the presence of specific cells and cell subtypes are indicative of a particular response to treatment, such as including increased or decreased susceptibility to treatment. The signature may indicate the presence of one particular cell type. In one embodiment, the novel signatures are used to detect multiple cell states or hierarchies that occur in subpopulations of cells that are linked to a particular pathological condition, or linked to a particular outcome or progression of the disease, or linked to a particular response to treatment of the disease.

[0052] The signature according to certain embodiments of the present invention may comprise or consist of one or more genes, proteins and / or epigenetic elements, such as for instance 1, 2, 3, 4, 5, 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of two or more genes, proteins and / or epigenetic elements, such as for instance 2, 3, 4, 5, 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of three or more genes, proteins and / or epigenetic elements, such as for instance 3, 4, 5, 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of four or more genes, proteins and / or epigenetic elements, such as for instance 4, 5, 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of five or more genes, proteins and / or epigenetic elements, such as for instance 5, 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of six or more genes, proteins and / or epigenetic elements, such as for instance 6, 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of seven or more genes, proteins and / or epigenetic elements, such as for instance 7, 8, 9, 10 or more. In certain embodiments, the signature may comprise or consist of eight or more genes, proteins and / or epigenetic elements, such as for instance 8, 9, 10 or more. Tn certain embodiments, the signature may comprise or consist of nine or more genes, proteins and / or epigenetic elements, such as for instance 9, 10 or more. In certain embodiments, the signature may comprise or consist often or more genes, proteins and / or epigenetic elements, such as for instance 10, 11, 12, 13, 14, 15, or more. It is to be understood that a signature according to the invention may for instance also include genes or proteins as well as epigenetic elements combined.

[0053] In certain embodiments, a signature is characterized as being specific for a population of cells that respond to a potential therapeutic ex vivo if it is upregulated or only present, detected or detectable in a population of cells that respond to a potential therapeutic ex vivo, or alternatively is downregulated or only absent, or undetectable in that population of cells that respond to a potential therapeutic ex vivo. In this context, a signature consists of one or more differentially expressed genes / proteins or differential epigenetic elements when comparing different cells or cell subpopulations, including comparing cells that do respond to a potential therapeutic ex vivo and cells that do not respond to a potential therapeutic ex vivo. It is to be understood that "differentially expressed" genes / proteins include genes / proteins which are up- or down-regulated as well as genes / proteins which are turned on or off. When referring to up- or downregulation, in certain embodiments, such up- or down-regulation is preferably at least two-fold, such as two-fold, three-fold, four-fold, five-fold, or more, such as for instance at least ten-fold, at least 20-fold, at least 30-fold, at least 40-fold, at least 50-fold, or more. Alternatively, or in addition, differential expression may be determined based on common statistical tests, as is known in the art.

[0054] As discussed herein, differentially expressed genes / proteins, or differential epigenetic elements may be differentially expressed on a single cell level, or may be differentially expressed on a cell population level. Preferably, the differentially expressed genes / proteins or epigenetic elements as discussed herein, such as constituting the gene signatures as discussed herein, when as to the cell population level, refer to genes that are differentially expressed in all or substantially all cells of the population (such as at least 80%, preferably at least 90%, such as at least 95% of the individual cells). This allows one to define a particular subpopulation of tumor cells. As referred to herein, a "subpopulation" of cells preferably refers to a particular subset of cells of a particular cell type which can be distinguished or are uniquely identifiable and set apart from other cells of this cell type. The cell subpopulation may be phenotypically characterized, and is preferably characterized by the signature as discussed herein. A cell (sub) population as referred to herein may constitute a subpopulation of cells of a particular cell type characterized by a specific cell state.

[0055] Various aspects and embodiments of the invention may involve analyzing gene signatures, protein signature, and / or other genetic or epigenetic signature based on single cell analyses (e.g., single cell RNA sequencing) or alternatively based on cell population or bulk analyses, as is defined herein elsewhere. In further aspects, the invention relates to gene signatures, protein signature, and / or other genetic or epigenetic signature of cells that are determined to respond to a therapeutic ex vivo, as defined herein elsewhere, and may therefore predict response to the therapeutic clinically in the patient.

[0056] The systems and methods provided herein further relate to various uses of the gene signatures, protein signature, and / or other genetic or epigenetic signature as defined herein, as well as various uses of the disease endotypes as defined herein. Particular advantageous uses include methods for identifying therapeutics predicted to be effective for a disease endotype based on the gene signatures, protein signature, and / or other genetic or epigenetic signature as defined herein. The systems and methods provided herein further relate to identifying therapeutics capable of treating patients assigned to a particular disease endotype based on the gene signatures, protein signature, and / or other genetic or epigenetic signature as defined herein.

[0057] The gene signatures described herein are useful in assigning a patient determined to have a particular disease to a particular endotype of that disease. The gene signatures described herein are further useful in predicting a patient’s response to a potential therapeutic for a particular disease endotype based on the patient being determined to have particular endotype of the disease.

[0058] In some embodiments, disease endotypes are determined from gene expression data derived from multiple datasets. In some embodiments, the gene expression data derived from multiple datasets are publicly available. In some embodiments, disease endotypes are determined from 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more datasets (see, FIG. 3A). In some embodiments, systematic disease endotype matching is performed across the multiple datasets (see, FIG. 3B). In some embodiments, each disease endotype is characterized by its gene expression signature (see, FIG. 3C).

[0059] In some embodiments, 1, 2, 3, 4, 5, 6, 7, 8, 9, or 10 or more disease endotypes are determined from gene expression data derived from multiple datasets, and a subset of disease endotypes are validated across the multiple datasets to confirm the validity of the disease endotypes. For example, four disease endotypes may be determined from gene expression data derived from multiple datasets, and two of the four disease endotypes may be validated across all the datasets (see, FIG. 4A). In some embodiments, expression signatures for individual genes may be compared across the multiple datasets in order to validate the determined disease endotypes. For example, a first subset of genes for which up-regulation is correlated with assigning a gene expression signature to a first disease endotype may be compared across multiple datasets to validate the determined disease endotype, and a second subset of genes for which down regulation is correlated with assigning a gene expression signature to a second disease endotype may be compared across multiple datasets to validate the determined disease endotype (see, FIG. 4B)

[0060] In certain embodiments, the systems and methods provided herein involve targeted nucleic acid profiling ( .g., sequencing, quantitative reverse transcription polymerase chain reaction, and the like). In certain embodiments, a target nucleic acid molecule (e.g., RNA molecule), may be sequenced by any method known in the art, for example, methods of high-throughput sequencing, also known as next generation sequencing or deep sequencing. A nucleic acid target molecule labeled with a barcode (for example, an origin-specific barcode) can be sequenced with the barcode to produce a single read and / or contig containing the sequence, or portions thereof, of both the target molecule and the barcode. Exemplary next generation sequencing technologies include, for example, Illumina sequencing, Ion Torrent sequencing, 454 sequencing, SOLiD sequencing, and nanopore sequencing amongst others.

[0061] In some embodiments, targeted nucleic acid profiling is performed on a biological sample obtained from a subject. The biological sample can be, for example, a peripheral blood sample, a urine sample, a saliva sample, a cerebrospinal fluid (CSF) sample, or a tissue biopsy. A biopsy can be, for example, a needle biopsy, fine-needle aspiration, core needle biopsy, vacuum-assisted biopsy, image-guide biopsy, endoscopic biopsy, skin biopsy, bone barrow biopsy, or surgical biopsy.

[0062] Tissues and cells types for which the methods disclosed herein can be used to identify disease subtypes based on gene expression data include, but are not limited to, peripheral blood mononuclear cells (PBMCs) including T cells and NK cells (TNK), monocytes and macrophages (MPh), B and plasma cells (BPC), neutrophils (Neut), erythrocytes (Eryth), megakaryocytes (Mega), and hematopoietic stem cells (HSC); macrophages and monocytes including classical (CD14+, CD16-), intermediate (CD14+, CD16+), and non-classical (CD14-, CD16+) monocytes; B cells including naive B cells (B.Naive), memory B cells (MBCs), and plasma cells (PCs); heart tissue, kidney tissue, and lung tissue including cardiac fibroblasts (CFs), cardiac muscle cells (CMC), endothelial cells (Endo), hepatocytes (Hepa), macrophages (Mac), and smooth muscle cells (SMC); kidney tissue including B cells (B), endothelial cells (Endo), Epithelial cells (Epit), macrophages (Mac), and T and natural killer cell population (TNKs); and lung tissue including adventitial cells (AC), alveolar type 1 and type 2 cells (ATC), basal cells (Basal), basophils (Baso), B and Plasma cells (BPC), club cells (CC), endothelial (Endo), fibroblasts (Fibro), goblet, serous, and mucous cells (GSM), lung ciliated cells (LCC), mesothelial cells (MC), monocytes, dendritic cells, and macrophages (MPh), neutrophils (Neut), pericyte cells (PC), pulmonary ionocytes (PI), smooth muscle cells (SMC), T and natural killer cells (TNK).

[0063] In some embodiments, the tissue for which the methods disclosed herein can be used to identify disease subtypes based on gene expression data include, but are not limited to prostate, lung, pancreas, cervical, renal, salivary gland uterine, gastric, thyroid, sinus, middle and inner ear, adrenal glands, appendix, hematopoietic system, bones and joints, spinal cord, breast, cerebellum, connective and soft tissue, corpus uteri, esophagus, eye, nose, eyeball, fallopian tube, extrahepatic bile ducts, mouth, intrahepatic bile ducts, kidney, appendix, larynx, lip, liver, lung and bronchus, lymph node, cerebral, spinal, nasal cartilage, retina, oropharynx, endocrine glands, female genital, ovary, penis and scrotum, pituitary gland, pleura, rectum, renal pelvis, ureter, peritoneum, salivary gland, skin, small intestine, testis, thymus, thyroid gland, tongue, unknown, urinary bladder, uterus, vagina, labia, or vulva tissue. In some embodiments, the sample comprises cells selected from the group consisting of adipose, adrenal cortex, adrenal gland, adrenal gland - medulla, appendix, bladder, blood, blood vessel, bone, bone cartilage, brain, breast, cartilage, cervix, colon, colon sigmoid, dendritic cells, skeletal muscle, endometrium, esophagus, fallopian tube, fibroblast, gallbladder, kidney, larynx, liver, lung, lymph node, melanocytes, mesothelial lining, myoepithelial cells, osteoblasts, ovary, pancreas, parotid, prostate, salivary gland, sinus tissue, skeletal muscle, skin, small intestine, smooth muscle, stomach, synovium, j oint lining tissue, tendon, testis, thymus, thyroid, uterus, and uterus corpus. In some embodiments, the tissue is collected from a healthy subject. In some embodiments, the tissue is collected from a subject who is known, or is suspected to have, a cancer. In some embodiments, the tissue is collected from a solid or liquid tumor.

[0064] In some embodiments, the diseases for which the methods disclosed herein can be used to identify disease subtypes based on gene expression data include, but are not limited to autoimmune myocarditis, anti-glomerular basement membrane nephritis, lupus nephritis, interstitial cystitis, autoimmune hepatitis, primary biliary cholangitis, primary sclerosing cholangitis, anti synthetase syndrome, alopecia areata, autoimmune angioedema, autoimmune progesterone dermatitis, autotimmune urticaria, bullous pemphigoid, cicatricial pemphigoid, dermatitis herpetiformis, discoid lupus erythematosus, epidermolysis bullosa acquisita, erythema nodosum, gestational pemphigoid, hidradenitis suppurativa, lichen planus, lichen sclerosus, linear IgA disease, mophea, pemphigus vulgaris, psoriasis, systemic scleroderma, vitiligo, Addison’s disease, autoimmune poly endocrine syndrome type 1, autoimmune poly endocrine syndrome type 2, autoimmune polyendocrine syndrome type 3, autoimmune pancreatitis, diabetes mellitus type 1, autoimmune thyroiditis, Ord’s thyroiditis, Graves’ disease, autoimmune oophoritis, endometriosis, autoimmune orchitis, autoimmune enteropathy, coeliac disease, Crohn’s disease, esophageal achalasia, microscopic colitis, ulcerative colitis, antiphospholoipid syndrome, aplastic anemia, autoimmune meolytic anemia, autoimmune lymphoproliferative syndrome, autoimmne neutropenia, autoimmune thrombocytopenic purpura, cold agglutinin disease, essential mixed cryoglobulinemia, Evans syndrome, pernicious anemia, pure red cell aplasia, thrombocytopenia, adiposis dolorosa, Adult-onset Still's disease, Ankylosing spondylitis, CREST syndrome, Drug- induced lupus, enthesitis-related arthritis, eosinophilic fasciitis, Felty syndrome, IgG4- related disease, juvenile arthritis, mixed connective tissue disease (MCTD), palindromic rheumatism, Parsonage-Turner syndrome, relapsing polychondritis, retroperitoneal fibrosis, rheumatic fever, rheumatoid arthritis, sarcoidosis, Schnitzler syndrome, systemic lupus erythematosus (SLE), undifferentiated connective tissue disease (UCTD), dermatomyositis, fibromyalgia, inclusion body myositis, myositis, myasthenia gravis, neuromyotonia, paraneoplastic cerebellar degeneration, polymyositis, Acute disseminated encephalomyelitis (ADEM) Acute motor axonal neuropathy, Anti-N-Methyl-D- Aspartate (Anti-NMDA) Receptor Encephalitis, Balo concentric sclerosis, Bickerstaffs encephalitis, chronic inflammatory demyelinating polyneuropathy, Guillain-Barre syndrome, Hashimoto's encephalopathy, idiopathic inflammatory demyelinating diseases, Lambert-Eaton myasthenic syndrome, multiple sclerosis, Oshtoran syndrome, progressive inflammatory neuropathy, restless legs syndrome, stiff-person syndrome, Sydenham's chorea, transverse myelitis, Autoimmune retinopathy, autoimmune uveitis, Cogan syndrome, Graves' ophthalmopathy, intermediate uveitis, ligneous conjunctivitis, Mooren's ulcer, neuromyelitis optica, opsoclonus myoclonus syndrome, optic neuritis, scleritis, Susac's syndrome, sympathetic ophthalmia, Tolosa-Hunt syndrome, autoimmune inner ear disease (AIED), Meniere's disease, Behcet's disease, eosinophilic granulomatosis with polyangiitis (EGPA), giant cell arteritis, granulomatosis with polyangiitis (GPA), IgA vasculitis (IgAV), Kawasaki disease, leukocytoclastic vasculitis, lupus vasculitis, rheumatoid vasculitis, microscopic polyangiitis (MPA), polyarteritis nodosa (PAN), polymyalgia rheumatic, urticarial vasculitis, vasculitis, and primary immunodeficiency.

[0065] Method of Treatment

[0066] It will be understood by the skilled person that treating as referred to herein encompasses enhancing treatment or improving treatment efficacy. Treatment may include ameliorating symptoms of the disease, slowing, halting, or reversing progression of the disease, or inhibition or reduction of otherwise deleterious effects associated with the disease. Efficaciousness of treatment is determined in association with any known method for diagnosing or treating the particular disease. The systems and methods provided herein comprehend a treatment method comprising any one of the methods or uses herein discussed.

[0067] The phrase "therapeutically effective amount" as used herein refers to a nontoxic but sufficient amount of a drug, agent, or compound to provide a desired therapeutic effect. As used herein "patient" refers to any human being receiving or who may receive medical treatment. Therapy or treatment according to the invention may be performed alone or in conjunction with another therapy, and may be provided at home, the doctor's office, a clinic, a hospital's outpatient department, or a hospital. Treatment generally begins at a hospital so that the doctor can observe the therapy's effects closely and make any adjustments that are needed. The duration of the therapy depends on the age and condition of the patient, the stage of the disease, and how the patient responds to the treatment. Additionally, a person having a greater risk of developing a disease (e.g., a person who is genetically predisposed to a disease) may receive prophylactic treatment to inhibit or delay symptoms of the disease.

[0068] Computer Implementation of the Methods

[0069] FIG. 5 is a diagram of computer system 500 components that can be used to implement methods for identifying disease subtypes based on gene expression data, for example, RNA-seq data or gene expression array data.

[0070] Computing device 500 is intended to represent various forms of digital computers, such as laptops, desktops, workstations, personal digital assistants, servers, blade servers, mainframes, and other appropriate computers. Computing device 550 is intended to represent various forms of mobile devices, such as personal digital assistants, cellular telephones, smartphones, and other similar computing devices. Additionally, computing device 500 or 550 can include Universal Serial Bus (USB) flash drives. The USB flash drives can store operating systems and other applications. The USB flash drives can include input / output components, such as a wireless transmitter or USB connector that can be inserted into a USB port of another computing device. The components shown here, their connections and relationships, and their functions, are meant to be examples only, and are not meant to limit implementations of the methods and compositions described and / or claimed in this document.

[0071] Computing device 500 includes a processor 502, memory 504, a storage device 506, a high-speed interface 508 connecting to memory 504 and high-speed expansion ports 510, and a low speed interface 512 connecting to low speed bus 514 and storage device 506. Each of the components 502, 504, 506, 508, 510, and 512, are interconnected using various busses, and can be mounted on a common motherboard or in other manners as appropriate. The processor 502 can process instructions for execution within the computing device 500, including instructions stored in the memory 504 or on the storage device 506 to display graphical information for a GUI on an external input / output device, such as display 516 coupled to high speed interface 508. In other implementations, multiple processors and / or multiple buses can be used, as appropriate, along with multiple memories and types of memory. Also, multiple computing devices 500 can be connected, with each device providing portions of the necessary operations, e.g., as a server bank, a group of blade servers, or a multi-processor system.

[0072] The memory 504 stores information within the computing device 500. In one implementation, the memory 504 is a volatile memory unit or units. In another implementation, the memory 504 is a non-volatile memory unit or units. The memory 504 can also be another form of computer-readable medium, such as a magnetic or optical disk.

[0073] The storage device 506 is capable of providing mass storage for the computing device 500. In one implementation, the storage device 506 can be or contain a computer- readable medium, such as a floppy disk device, a hard disk device, an optical disk device, or a tape device, a flash memory or other similar solid-state memory device, or an array of devices, including devices in a storage area network or other configurations. A computer program product can be tangibly embodied in an information carrier. The computer program product can also contain instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer- or machine-readable medium, such as the memory 504, the storage device 506, or memory on processor 502. The high-speed controller 508 manages bandwidth -intensive operations for the computing device 500, while the low speed controller 512 manages lower bandwidth intensive operations. Such allocation of functions is only an example. In one implementation, the high-speed controller 508 is coupled to memory 504, display 516, e.g., through a graphics processor or accelerator, and to high-speed expansion ports 510, which can accept various expansion cards (not shown). In the implementation, low-speed controller 512 is coupled to storage device 506 and low-speed expansion port 514. The low-speed expansion port, which can include various communication ports, e.g., USB, Bluetooth, Ethernet, wireless Ethernet can be coupled to one or more input / output devices, such as a keyboard, a pointing device, mi crophone / speaker pair, a scanner, or a networking device such as a switch or router, e.g., through a network adapter. The computing device 500 can be implemented in a number of different forms, as shown in the figure. For example, it can be implemented as a standard server 520, or multiple times in a group of such servers. It can also be implemented as part of a rack server system 524. In addition, it can be implemented in a personal computer such as a laptop computer 522. Alternatively, components from computing device 500 can be combined with other components in a mobile device (not shown), such as device 550. Each of such devices can contain one or more of computing device 500, 550, and an entire system can be made up of multiple computing devices 500, 550 communicating with each other.

[0074] The computing device 500 can be implemented in a number of different forms, as shown in the figure. For example, it can be implemented as a standard server 520, or multiple times in a group of such servers. It can also be implemented as part of a rack server system 524. In addition, it can be implemented in a personal computer such as a laptop computer 522. Alternatively, components from computing device 500 can be combined with other components in a mobile device (not shown), such as device 550. Each of such devices can contain one or more of computing device 500, 550, and an entire system can be made up of multiple computing devices 500, 550 communicating with each other.

[0075] Computing device 550 includes a processor 552, memory 564, and an input / output device such as a display 554, a communication interface 566, and a transceiver 568, among other components. The device 550 can also be provided with a storage device, such as a micro-drive or other device, to provide additional storage. Each of the components 550, 552, 564, 554, 566, and 568, are interconnected using various buses, and several of the components can be mounted on a common motherboard or in other manners as appropriate.

[0076] The processor 552 can execute instructions within the computing device 550, including instructions stored in the memory 564. The processor can be implemented as a chipset of chips that include separate and multiple analog and digital processors. Additionally, the processor can be implemented using any of a number of architectures. For example, the processor 510 can be a CISC (Complex Instruction Set Computers) processor, a RISC (Reduced Instruction Set Computer) processor, or a MISC (Minimal Instruction Set Computer) processor. The processor can provide, for example, for coordination of the other components of the device 550, such as control of user interfaces, applications run by device 550, and wireless communication by device 550.

[0077] Processor 552 can communicate with a user through control interface 558 and display interface 556 coupled to a display 554. The display 554 can be, for example, a TFT (Thin-Film-Transistor Liquid Crystal Display) display or an OLED (Organic Light Emitting Diode) display, or other appropriate display technology. The display interface 556 can comprise appropriate circuitry for driving the display 554 to present graphical and other information to a user. The control interface 558 can receive commands from a user and convert them for submission to the processor 552. In addition, an external interface 562 can be provided in communication with processor 552, so as to enable near area communication of device 550 with other devices. External interface 562 can provide, for example, for wired communication in some implementations, or for wireless communication in other implementations, and multiple interfaces can also be used. The memory 564 stores information within the computing device 550. The memory 564 can be implemented as one or more of a computer-readable medium or media, a volatile memory unit or units, or a non-volatile memory unit or units. Expansion memory 574 can also be provided and connected to device 550 through expansion interface 572, which can include, for example, a SIMM (Single In Line Memory Module) card interface. Such expansion memory 574 can provide extra storage space for device 550, or can also store applications or other information for device 550. Specifically, expansion memory 574 can include instructions to carry out or supplement the processes described above, and can also include secure information. Thus, for example, expansion memory 574 can be provided as a security module for device 550, and can be programmed with instructions that permit secure use of device 550. In addition, secure applications can be provided via the SIMM cards, along with additional information, such as placing identifying information on the SIMM card in a non-hackable manner.

[0078] The memory can include, for example, flash memory and / or NVRAM memory, as discussed below. In one implementation, a computer program product is tangibly embodied in an information carrier. The computer program product contains instructions that, when executed, perform one or more methods, such as those described above. The information carrier is a computer- or machine-readable medium, such as the memory 564, expansion memory 574, or memory on processor 552 that can be received, for example, over transceiver 568 or external interface 562.

[0079] Device 550 can communicate wirelessly through communication interface 566, which can include digital signal processing circuitry where necessary. Communication interface 566 can provide for communications under various modes or protocols, such as GSM voice calls, SMS, EMS, or MMS messaging, CDMA, TDMA, PDC, WCDMA, CDMA2000, or GPRS, among others. Such communication can occur, for example, through radio-frequency transceiver 568. In addition, short-range communication can occur, such as using a Bluetooth, Wi-Fi, or other such transceiver (not shown). In addition, GPS (Global Positioning System) receiver module 570 can provide additional navigation- and location-related wireless data to device 550, which can be used as appropriate by applications running on device 550.

[0080] Device 550 can also communicate audibly using audio codec 560, which can receive spoken information from a user and convert it to usable digital information. Audio codec 560 can likewise generate audible sound for a user, such as through a speaker, e.g., in a handset of device 550. Such sound can include sound from voice telephone calls, can include recorded sound, e.g., voice messages, music files, etc. and can also include sound generated by applications operating on device 550.

[0081] The computing device 550 can be implemented in a number of different forms, as shown in the figure. For example, it can be implemented as a cellular telephone 580. It can also be implemented as part of a smartphone 582, personal digital assistant, or other similar mobile device.

[0082] Various implementations of the systems and methods described here can be realized in digital electronic circuitry, integrated circuitry, specially designed ASICs (application specific integrated circuits), computer hardware, firmware, software, and / or combinations of such implementations. These various implementations can include implementation in one or more computer programs that are executable and / or interpretable on a programmable system including at least one programmable processor, which can be special or general purpose, coupled to receive data and instructions from, and to transmit data and instructions to, a storage system, at least one input device, and at least one output device.

[0083] These computer programs (also known as programs, software, software applications or code) include machine instructions for a programmable processor, and can be implemented in a high-level procedural and / or object-oriented programming language, and / or in assembly / machine language. As used herein, the terms "machine-readable medium" "computer-readable medium" refers to any computer program product, apparatus and / or device, e.g., magnetic discs, optical disks, memory, Programmable Logic Devices (PLDs), used to provide machine instructions and / or data to a programmable processor, including a machine-readable medium that receives machine instructions as a machine-readable signal. The term "machine-readable signal" refers to any signal used to provide machine instructions and / or data to a programmable processor. To provide for interaction with a user, the systems and techniques described here can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input. The systems and techniques described here can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the systems and techniques described here, or any combination of such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network ("LAN"), a wide area network ("WAN"), and the Internet.

[0084] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other. A number of embodiments have been described. Nevertheless, it will be understood that various modifications can be made without departing from the spirit and scope of the invention. In addition, the logic flows depicted in the figures do not require the particular order shown, or sequential order, to achieve desirable results. In addition, other steps can be provided, or steps can be eliminated, from the described flows, and other components can be added to, or removed from, the described systems. Accordingly, other embodiments are within the scope of the following claims.

[0085] Embodiments of the disclosure and all of the functional operations described in this specification can be implemented in digital electronic circuitry, or in computer software, firmware, or hardware, including the structures disclosed in this specification and their structural equivalents, or in combinations of one or more of them. Embodiments of the methods and compositions can be implemented as one or more computer program products, e.g., one or more modules of computer program instructions encoded on a computer readable medium for execution by, or to control the operation of, data processing apparatus. The computer readable medium can be a machine-readable storage device, a machine-readable storage substrate, a memory device, a composition of matter effecting a machine-readable propagated signal, or a combination of one or more of them. The term “data processing apparatus” encompasses all apparatus, devices, and machines for processing data, including by way of example a programmable processor, a computer, or multiple processors or computers. The apparatus can include, in addition to hardware, code that creates an execution environment for the computer program in question, e.g., code that constitutes processor firmware, a protocol stack, a database management system, an operating system, or a combination of one or more of them. A propagated signal is an artificially generated signal, e.g., a machine-generated electrical, optical, or electromagnetic signal that is generated to encode information for transmission to suitable receiver apparatus.

[0086] A computer program (also known as a program, software, software application, script, or code) can be written in any form of programming language, including compiled or interpreted languages, and it can be deployed in any form, including as a stand alone program or as a module, component, subroutine, or other unit suitable for use in a computing environment. A computer program does not necessarily correspond to a file in a file system. A program can be stored in a portion of a file that holds other programs or data (e.g., one or more scripts stored in a markup language document), in a single file dedicated to the program in question, or in multiple coordinated files (e.g., files that store one or more modules, sub programs, or portions of code). A computer program can be deployed to be executed on one computer or on multiple computers that are located at one site or distributed across multiple sites and interconnected by a communication network.

[0087] The processes and logic flows described in this specification can be performed by one or more programmable processors executing one or more computer programs to perform functions by operating on input data and generating output. The processes and logic flows can also be performed by, and apparatus can also be implemented as, special purpose logic circuitry, e.g., an FPGA (field programmable gate array) or an ASIC (application specific integrated circuit).

[0088] Processors suitable for the execution of a computer program include, by way of example, both general and special purpose microprocessors, and any one or more processors of any kind of digital computer. Generally, a processor will receive instructions and data from a read only memory or a random access memory or both. The essential elements of a computer are a processor for performing instructions and one or more memory devices for storing instructions and data. Generally, a computer will also include, or be operatively coupled to receive data from or transfer data to, or both, one or more mass storage devices for storing data, e.g., magnetic, magneto optical disks, or optical disks. However, a computer need not have such devices. Moreover, a computer can be embedded in another device, e.g., a tablet computer, a mobile telephone, a personal digital assistant (PDA), a mobile audio player, a Global Positioning System (GPS) receiver, to name just a few. Computer readable media suitable for storing computer program instructions and data include all forms of non volatile memory, media and memory devices, including by way of example semiconductor memory devices, e.g., EPROM, EEPROM, and flash memory devices; magnetic disks, e.g., internal hard disks or removable disks; magneto optical disks; and CD ROM and DVD-ROM disks. The processor and the memory can be supplemented by, or incorporated in, special purpose logic circuitry.

[0089] To provide for interaction with a user, embodiments of the disclosure can be implemented on a computer having a display device, e.g., a CRT (cathode ray tube) or LCD (liquid crystal display) monitor, for displaying information to the user and a keyboard and a pointing device, e.g., a mouse or a trackball, by which the user can provide input to the computer. Other kinds of devices can be used to provide for interaction with a user as well; for example, feedback provided to the user can be any form of sensory feedback, e.g., visual feedback, auditory feedback, or tactile feedback; and input from the user can be received in any form, including acoustic, speech, or tactile input.

[0090] Embodiments of the disclosure can be implemented in a computing system that includes a back end component, e.g., as a data server, or that includes a middleware component, e.g., an application server, or that includes a front end component, e.g., a client computer having a graphical user interface or a Web browser through which a user can interact with an implementation of the methods, or any combination of one or more such back end, middleware, or front end components. The components of the system can be interconnected by any form or medium of digital data communication, e.g., a communication network. Examples of communication networks include a local area network (“LAN”) and a wide area network (“WAN”), e.g., the Internet.

[0091] The computing system can include clients and servers. A client and server are generally remote from each other and typically interact through a communication network. The relationship of client and server arises by virtue of computer programs running on the respective computers and having a client-server relationship to each other.

[0092] While this specification contains many specifics, these should not be construed as limitations on the scope of the invention or of what may be claimed, but rather as descriptions of features specific to particular embodiments of the invention. Certain features that are described in this specification in the context of separate embodiments can also be implemented in combination in a single embodiment. Conversely, various features that are described in the context of a single embodiment can also be implemented in multiple embodiments separately or in any suitable subcombination. Moreover, although features may be described above as acting in certain combinations and even initially claimed as such, one or more features from a claimed combination can in some cases be excised from the combination, and the claimed combination may be directed to a subcombination or variation of a subcombination.

[0093] Similarly, while operations are depicted in the drawings in a particular order, this should not be understood as requiring that such operations be performed in the particular order shown or in sequential order, or that all illustrated operations be performed, to achieve desirable results. In certain circumstances, multitasking and parallel processing may be advantageous. Moreover, the separation of various system components in the embodiments described above should not be understood as requiring such separation in all embodiments, and it should be understood that the described program components and systems can generally be integrated together in a single software product or packaged into multiple software products.

[0094] In each instance where an HTML file is mentioned, other file types or formats may be substituted. For instance, an HTML file may be replaced by an XML, JSON, plain text, or other types of files. Moreover, where a table or hash table is mentioned, other data structures (such as spreadsheets, relational databases, or structured files) may be used. Example 1 -

[0095] The approach depicted in Figure 1A-2B highlights four stages of the methods and systems disclosed herein:

[0096] • Stage 1 : Generation of a pathway-activation signature through genomewide transcriptome profiling of ex vivo or in vitro cell models treated with, for example, pathway activating agents and a drug that blocks the pathway (FIG. 1A).

[0097] • Stage 2: Identification of reproducible disease endotypes across multiple study cohorts (FIG. IB).

[0098] • Stage 3: Evaluation of the pathway activation signature across the disease endotypes identified in Stage 2 (FIG. 2A).

[0099] • Stage 4: Development of a molecular classifier for the disease endotypes prioritized in Stage 3 to inform clinical trial design (FIG. 2B).

[0100] These four stages are described in details below.

[0101] Stage 1: Generation of a drug response gene signature

[0102] To generate a drug response gene signature, genome-wide expression profiling of ex vivo or in vitro models treated with a drug are used to identify pharmacodynamic markers that are hypothesized to indicate drug responsiveness. An experimental design is shown in Table 1 below. Samples treated at different timepoints are profiled to identify genes that are consistently up- or down-regulated over time.

[0103] Table 1. Study design of a genomic profiling study for drug signature gene identification

[0104] The drug signature is defined as genes that are consistently up-regulated under ligand stimulation / pathway activation and the expression changes are reversed by the addition of the drug (FIG. 1A). Statistical modeling approaches such as linear mixed effect (LME) modeling are used to assess how likely a gene is impacted by ligand stimulation and drug treatment while accounting for individual donor variation. An FDR threshold of 0.05 with effect size (fold change) threshold of 1.5 is used to filter genes to generate the drug signature. Finally, additional analyses such as pathway enrichment are conducted to evaluate the drug gene signature with respect to pathways reported in the published literature.

[0105] Stage 2: Identification of reproducible disease endotypes based on whole-genome expression profiling of patient samples

[0106] For a given disease a collection of transcriptome profiling data are generated from multiple cohorts, and these multiple cohorts are used to evaluate the reproducibility of the identified disease endotypes. In addition to transcriptomic profiling, data such as patient demographics, clinical measurements and treatment responses are used to inform the characterization of different disease endotypes. A data-driven approach is applied for the identification and characterization of disease endotypes. This approach includes three steps:

[0107] Step 1 - Unbiased clustering analysis to identify patient clusters in each cohort. (FIG. 3 A)

[0108] Starting with transcriptome profiling data of disease tissue, a k-means clustering analysis coupled with GAP statistic is used to identify subgroups of patients within each cohort. The number of clusters (K) for each cohort is determined with GAP statistic. Each dataset (or cohort) is analyzed independently to identify subgroups of patients from each dataset. Step 2 - Systematic subgroup matching across multiple cohorts for the identification of reproducible disease endotypes. (FIG. 3B)

[0109] To match patient subgroups from different study cohorts, a systematic comparison between all subgroups from all cohorts is performed and subgroups from each cohort are grouped into aggregated subtypes (i.e., endotypes) that are stable and consistently present in multiple cohorts. This systemic comparison is performed by looking at the concordance or correlation of subgroup specific gene expression changes. Subgroups of patients from different cohorts are considered as a match (with similar gene expression characteristics) and grouped into endotypes when they are reciprocally the best hit with Pearson correlation coefficient greater than 0.7. An example of finding consistent disease endotypes across multiple cohorts is shown in FIG. 4A. In this example, subgroups A and B of FIG. 4A are consistently present in all study cohorts and considered as candidate disease endotypes.

[0110] Step 3 - Characterization of candidate disease endotypes (FIG. 3C)

[0111] Once disease endotypes are identified across multiple study cohorts, they are further characterized by associated clinical features, multiple gene modules (for example, relating to tissue morphology, cellular functions and pathways) and response to standard of care therapies as shown in FIG. 4B.

[0112] Stage 3: Evaluation of the drug response signature across the disease endotypes

[0113] With disease endotypes consistently present in multiple study cohorts, a connection is established between disease endotypes and response to the treatment of interest to generate hypotheses for targeted clinical therapeutics. It is hypothesized that patients with higher activities of the drug response gene signature are more likely to be better responders to the specific therapy. In the example of FIG. 4B, endotype A is associated with higher activity of the drug response signature and is predicted to be the target patient population for the specific treatment.

[0114] Stage 4: Translating endotypes into the clinic - development of molecular endotype classifiers to enable targeted therapies In the example of FIG. 4B, patients of endotype A are associated with higher drug response gene signature activity and are suggested as the target population for the specific therapy of interest. A molecular classifier can identify this target patient population based on gene expression data from the disease-relevant tissue. As shown in FIG. 6, a classifier is built with one study cohort as a training dataset (which includes parameter optimization and feature selection), then the classifier is applied to the testing datasets which include the remaining cohorts independent of the training data. The performance of the classifier is evaluated with metrics such as Cohen’s kappa, sensitivity, specificity, and confusion matrix. Algorithms such as AdaBoost (Adaptive Boosting), elastic net, SVM or other machine-learning methods are employed for the classifier development. Cross-validation and re-sampling approaches are applied to help select the optimal hyperparameters and gene features for the classifier.

[0115] By utilizing assay platforms such as qPCR or NanoString to quantify gene expression data from patients, it is possible to leverage the classifier to assign each patient with a disease endotype label. This predictive capability enables healthcare professionals to administer the most suitable treatment for the patient based on the treatment response hypotheses developed in Stage 3 of the framework.

[0116] OTHER EMBODIMENTS

[0117] It is to be understood that while the invention has been described in conjunction with the detailed description thereof, the foregoing description is intended to illustrate and not limit the scope of the invention, which is defined by the scope of the appended claims. Other aspects, advantages, and modifications are within the scope of the following claims.

Claims

WHAT IS CLAIMED IS:

1. A method of determining a disease endotype for a subject having the disease, the method comprising: defining, based on gene expression data for a plurality of target genes, one or more gene expression signatures corresponding to one or more endotypes of the disease; measuring expression levels of each of a plurality of genes in a biological sample from the subject; generating, based on expression levels of each of the plurality of genes in the biological sample from the subject, a gene expression signature for the subject; comparing the gene expression signature for the subject to the one or more gene expression signatures corresponding to one or more endotypes of the disease; and assigning, based on the comparison of gene expression signatures, a disease endotype to the subject having the disease.

2. The method of claim 1, further comprising predicting a response to an exogenous treatment based on the disease endotype assigned to the subject having the disease.

3. The method of any one of claims 1 or 2, wherein the gene expression data is RNA-seq data.

4. The method of any one of claims 1-3, wherein the gene expression data for a plurality of target genes is from two or more previously generated datasets.

5. The method of any one of claims 1-4, wherein the defining step further includes use of a machine learning algorithm.

6. The method of claim 5, wherein the machine learning algorithm further includes K-means clustering.

7. The method of any one of claims 1-6, wherein generating the gene expression signature utilizes K-means clustering.

8. The method of any one of claims 1-7, wherein the disease is a cancer.

9. The method of any one of claims 1-8, wherein the disease is an autoimmune disease.

10. The method of any one of claims 1-9, wherein the biological sample is a blood sample.

11. The method of any one of claims 1-10, wherein the biological sample is a urine sample.

12. A method of treating a disease in a subject, the method comprising: obtaining an expression level of each of a plurality of genes in a biological sample obtained from the subject; assigning the disease of the subject to a disease endotype based on the expression level of each of the plurality of genes in the biological sample, the disease endotype being selected from a set of endotypes previously determined from gene expression data derived from multiple datasets; administering a therapeutic agent to the subject, wherein the assigning the disease of the subject to the disease endotype indicates that the therapeutic agent is predicted to be effective for treating the disease in the subject.

13. The method of claim 12, wherein the biological sample is a blood sample.

14. The method of claims 13, wherein the biological sample is a urine sample.

15. The method of any one of claims 12-14, wherein the level of expression of one or more of the plurality of genes of the disease endotype is increased relative to genes that are not assigned to the disease endotype.

16. The method of any one of claims 12-15, wherein the level of expression of one or more of the plurality of genes of the disease endotype is decreased relative to genes that are not assigned to the disease endotype.

17. A computer-implemented method for determining a disease endotype for a subject having the disease, the method comprising: receiving, by a computing device including a processor programmed to execute software instructions within a memory, a set of gene expression data comprising expression levels of each of a plurality of genes in a biological sample obtained from the subject; applying a machine learning algorithm to generate a classification model used in a classification scheme, wherein the machine learning algorithm has been trained using a stratified train-test split of the set of gene expression data; applying, by the computing device, the classification scheme, to rank the set of gene expression data; generating, by the computing device, a reduced set of genes that distinguish disease endotypes; determining, based on the expression levels of the reduced set of genes that distinguish disease endotypes, the disease endotype for the subject having the disease.

18. The computer-implemented method of claim 17, further comprising predicting a response to an exogenous treatment based on the disease endotype determined for the subject having the disease.

19. The computer-implemented method of claim 18, wherein the machine learning algorithm comprises feature selection.

20. The computer-implemented method of claim 19, wherein the feature selection is forward feature group selection (FFGS).

21. A computing system for determining a disease endotype for a subject having the disease, the computing system comprising: a data server for receiving gene expression data for a plurality of target genes and measurements of expression levels of each of a plurality of genes in a biological sample from the subject; a computing device communicatively connected to the data server, the computing device comprising an application server configured to: define, based on the gene expression data, one or more gene expression signatures corresponding to one or more endotypes of the disease; generate, based on the expression levels of each of the plurality of genes in the biological sample from the subject, a gene expression signature for the subject; compare the gene expression signature for the subject to the one or more gene expression signatures corresponding to one or more endotypes of the disease; and assign, based on the comparison of gene expression signatures, a disease endotype to the subject having the disease; and a display communicatively connected to the computing device and configured to display a report describing the assignment of the disease endotype to the subject having the disease.

22. The computing system of claim 21, wherein the application server is further configured to predict a response to an exogenous treatment based on the disease endotype assigned to the subject having the disease.

23. The computing system of any one of claims 21 or 22, wherein the gene expression data is RNA-seq data.

24. The computing system of any one of claims 21-23, wherein the gene expression data for a plurality of target genes is from two or more previously generated datasets.

25. The computing system of any one of claims 21-24, wherein the defining step further includes use of a machine learning algorithm.

26. The computing system of claim 25, wherein the machine learning algorithm further includes K-means clustering.

27. The computing system of any one of claims 21-26, wherein generating the gene expression signature utilizes K-means clustering.

28. The computing system of any one of claims 21-27, wherein the disease is a cancer.

29. The computing system of any one of claims 21-28, wherein the disease is an autoimmune disease.

30. The computing system of any one of claims 21-29, wherein the biological sample is a blood sample.

31. The computing system of any one of claims 21-30, wherein the biological sample is a urine sample.

Images