WO2017139756A1 - Methods of ascertaining cross-subject quantitative relationships between response indicators among multiple subjects in a population using multipotent or pluripotent stem cells - Google Patents

Abstract

Methods are described herein for ascertaining how biological responses of one type, or due to one stimulus, are statistically related to any of: (1) biological reactions of another type; or (2) biological reactions due to a different stimulus; or (3) any biological feature of the stimulus, across multiple test subjects of a population. Assays and other methods of measurement are conducted on stem cells, multipotent or pluripotent stem cells or cells differentiated therefrom, of multiple test subjects of a population. By comparing various endpoints from one or more assays for a single subject, an arithmetic relationship between measured endpoints is established. By comparing arithmetic relationships across all of subjects of a population, it can be ascertain whether such relationships are common across all subjects of the population, related through a pattern (such as linear or non-linear correlations), or idiosyncratic to each subject.

Description

METHODS OF ASCERTAINING CROSS-SUBJECT QUANTITATIVE RELATIONSHIPS BETWEEN RESPONSE INDICATORS AMONG MULTIPLE SUBJECTS IN A POPULATION USING MULTIPOTENT OR PLURIPOTENT

STEM CELLS

CROSS REFERENCE TO RELATED APPLICATIONS

[0001] This application claims the benefit of U.S. Provisional Application No.

62/294,101, filed on February 11, 2016, which is incorporated by reference herein in its entirety.

FIELD OF INVENTION

[0002] The present invention relates to the field of bio-analytics, especially methods for analyzing patterns of biological responses to a stimulus in a population.

BACKGROUND

[0003] Scientists have long been interested in the relationships among multiple biological responses to a single stimulus. For example, medical professionals have long relied on the relationships between various maladies such as disease or infection and a human's resulting fever in order to use temperature measurement as an indicator of whether the body is under certain stresses, such as infection.

[0004] Scientists have also long been interested in the relationships of a population's biological response(s) to one stimulus in comparison to the response(s) by that same population to other stimuli. An early example of such relationships was the demonstration in 1796 by Edward Jenner that infection by the (largely benign) cowpox virus could inoculate humans against the far more deadly smallpox virus.

[0005] Many such cross-subject and cross-reaction relationships have been studied and proven to be useful in the life sciences today.

[0006] As science has progressed, it is no longer sufficient to examine such questions in a binary form. For example, an investigator today would want to know not only that a person's exposure to cowpox might prevent infection by smallpox, but also whether differences in the severity of the cowpox infection correlated with the resilience of the cowpox "victim" to various levels of exposure to the smallpox virus. [0007] In recent years, scientists have become interested in such relationships at the cellular or tissue level. However, efforts to ascertain such cellular or tissue level

relationships have been difficult, due to the shortcomings of suitable test platforms.

[0008] In vivo observations face a variety of limitations. First, for obvious ethical reasons, in vivo observations are limited to benign stimuli, or to being ancillary to stimuli applied for other (presumably therapeutic) reasons. Second, researchers do not have the luxury of "repeating" experiments with incremental changes to the stimuli without suffering the potential for contamination of results from the residual effects of the prior iteration of the experiment. Third, examination of cellular or tissue level effects generally requires extraction of the cells or tissues involved - a process that may be itself unethical, and even if not, creates cost, contamination, risk of adverse consequences, and potential refusal by the test subjects.

[0009] The use of primary cells previously extracted from test subjects is likewise difficult, owing to the limited number of cells available from a single test subject and the inability of most such cells to be successfully cryopreserved and revived without significant alterations.

[0010] Thus, scientists have become interested in the use of stem cells and their derivatives. These cells provide a platform for non-invasive study that is infinitely replicable, and thus can theoretically support multiple experiments. However, to date, these platforms have not been developed beyond a very small number of parallel donor lines, owing to the difficulties associated with the potentially idiosyncratic behavior of (seemingly identically derived) cells from one person versus another.

[0011] Recently, scientists have laid the groundwork to compare the results of a single experiment conducted on cells from multiple individuals (and subsequently "averaged" or otherwise aggregated to form a representative view of a species' response) to the results of a second (or third, etc.) experiment (again, averaged or otherwise aggregated). Similarly, this same class of analysis may involve forming a relationship between the results of the two experiments at the level of the individual subject, but then averaging or aggregating those relationships into a single view of a species' response.

[0012] However, there is a second class of comparisons of multiple experiments - comparisons wherein the relationships among the experimental outcomes are formed at the individual subject level, but these individual relationships are compared and contrasted with each other to form a view about the correlation (broadly defined) of the degree of response in the first experiment and the degree of response in the second (or third, etc.) experiment. More formally, this class can be characterized via the following question: "When analyzed across the subjects of an experimental population, and without aggregating the results of individual subjects into any collective measure of the population (such as an average, etc.), what is the quantitative relationship among two or more designated measures of biological activity or features ("response indicators"; e.g., the scores on two endpoints of assays, or the scores of an assay contrasted with a biological feature such as size), referred to here as Measure A and Measure B? Such relationships can be:

[0013] Case 1 : Measure A and Measure B each exhibit a narrow range of values across the subjects within the population, but the value of Measure A in any subject is unrelated to the value of Measure B in that same subject;

[0014] Case 2: Either Measure A or Measure B (but not both) exhibits a wide range of values across subjects within the population, but the value of the other Measure in a subject is unrelated to the value of the wide-ranging Measure;

[0015] Case 3: Both Measure A and Measure B exhibit wide ranges of values across subjects within the population, but the value of Measure A in a subject is unrelated to the value of Measure B in that same subject;

[0016] Case 4: The value of Measure A in a subject bears a mathematical relationship to the value of Measure B in that same subject. This relationship can be of any mathematical form, including but not limited to: linear or non-linear correlations, step functions (e.g., a particular constant value of Measure A when Measure B lies within a specified range, but a different constant value of Measure A when Measure B lies within another range), etc.;

[0017] Case 5: Any combination of the above cases, such as when Case 4 exists for certain ranges of Measure A, but Case 1 exists in other ranges of Measure A." [0018] This class of relationships is more difficult to address than the first class for a variety of reasons, including the requirement for a larger number of cell lines to be developed, maintained and utilized in parallel.

[0019] Therefore, there remains a need for an improved method of ascertaining patterns of relationships between biological responses among multiple test subjects in a population when the preferred platform is cellular or tissue in nature, and the relationship to be investigated is cross-subj ect in nature.

SUMMARY

[0020] Methods are provided herein for ascertaining cross-subject quantitative relationships between biological reactions, or response indicators, among multiple subjects in a population using multipotent or pluripotent stem cells.

[0021] The methods involve the use of multiple assays and other methods of measurement conducted on cells of multiple subjects of a population, wherein the assays and/or measurements conducted on a subject are conducted on stem cells (or their derivatives), and the multiple experiments are conducted under protocols that hold most determinants of results constant across the various experiments, while varying only those determinants whose impact the research desires to directly compare.

[0022] By comparing various endpoints from one or more assays or measurements for a single subject, a mathematical relationship between the endpoints is established. By then comparing these mathematical relationships across all the subjects in a population, the researcher can ascertain whether such relationships are common across all subjects of the population, or whether the relationships are related through a pattern (such as linear or nonlinear correlations), or whether the mathematical relationships are idiosyncratic to each subject.

[0023] In some embodiments, methods for determining a relationship between two or more response indicators in a population comprise a) selecting a platform regimen comprising a set of variables that selects separate samples from a population of at least 10 subjects, wherein the samples comprise proliferative multipotent or pluripotent stem cells; b) selecting an experiment regimen comprising a set of controlled experimental conditions and at least one stimulus, wherein the at least one stimulus is applied to or combined with the sample to generate a response; c) assaying the samples for response to the at least one stimulus, wherein a response indicator comprises a quantified response endpoint to the at least one stimulus; d) determining for each sample an intra-subject relationship between two or more response indicators; and e) comparing in a cross-subject study the intra-subject relationships between the two or more response indicators to determine a cross-subject relationship between the two or more response indicators. In some embodiments, the methods further comprise differentiating the samples into one or more functional cell types. In some embodiments, the stimulus comprises an agent such as, but not limited to, a biological or chemical compound.

BRIEF DESCRIPTION OF THE FIGURES

[0024] The present invention may be better understood by referring to the following non- limiting figures.

[0025] Figure 1 is a schematic illustrating an example platform regimen. The regimen shows several example parameters, or variables, and the possible variants of each parameter that may be selected.

[0026] Figure 2 is a schematic illustrating an example experiment regimen. The regimen shows several example parameters, or variables, and the possible variants of each parameter that may be selected.

DETAILED DESCRIPTION

[0027] Titles or subtitles may be used in the specification for the convenience of a reader, which are not intended to influence the scope of the present invention.

Definitions

[0028] As will be understood by one skilled in the art, for any and all purposes, particularly in terms of providing a written description, all ranges disclosed herein also encompass any and all possible subranges and combinations of subranges thereof. Any listed range can be easily recognized as sufficiently describing and enabling the same range being broken down into at least equal halves, thirds, quarters, fifths, tenths, etc. As a non-limiting example, each range discussed herein can be readily broken down into a lower third, middle third and upper third, etc. It will be further understood that the endpoints of each of the ranges are significant both in relation to the other endpoint, and independently of the other endpoint. Thus, the endpoint of one range is combinable with the endpoint of another range. As will also be understood by one skilled in the art all language such as "up to," "at least," "greater than," "less than," and the like, include the number recited and refer to ranges which can be subsequently broken down into subranges as discussed above. Finally, as will be understood by one skilled in the art, a range includes each individual member. Thus, for example, a group having 1-3 samples refers to groups having 1 , 2, or 3 samples. Similarly, a group having 1-5 samples refers to groups having 1 , 2, 3, 4, or 5 samples, and so forth.

[0029] The following terms, unless otherwise indicated, shall be understood to have the following meanings:

[0030] As used herein, the terms "a," "an," and "the" include plural referents unless expressly and unequivocally limited to one referent.

[0031] The use of the term "or" is used to mean "and/or" unless explicitly indicated to refer to alternatives only or the alternatives are mutually exclusive, although the disclosure supports a definition that refers to only alternatives and "and/or." As used herein "another" can mean at least a second or more.

[0032] Throughout this application, the term "about" is used to indicate that a value includes the inherent variation of error for the device, the method being employed to determine the value, or the variation that exists among samples. It is to be understood, although not always explicitly stated, that all numerical designations may be preceded by the term "about."

[0033] The term "comprising" or "comprises" is intended to mean that the methods include the recited elements, but not excluding others. "Consisting essentially of when used to define methods, shall mean excluding other elements of any essential significance to the combination. For example, a method consisting essentially of the elements as defined herein would not exclude other elements that do not materially affect the basic and novel characteristic(s) of the claimed method. "Consisting of shall mean excluding more than trace amount of other ingredients and substantial method steps. Embodiments defined by each of these transition terms are within the scope of this invention.

[0034] The term "exposing" means contacting or placing in sufficiently close proximity at least one stimulus such as an agent (e.g. biological or chemical agent) or form of energy (e.g. radiation) to at least one object such as a sample (e.g. biological cells). The exposure typically, but need not necessarily, results in an observable or measurable biological or chemical phenomenon. The term is intended to encompass exposing at least a first object (e.g. fluorophore) in an inactive state to at least a second object (e.g. biological cell), and in which the first object becomes activated upon exposure of at least a third object (e.g.

chemical activator) or form of energy (e.g. radiation).

[0035] "Optional" or "optionally" means that the subsequently described event, circumstance, or material may or may not occur or be present, and that the description includes instances where the event, circumstance, or material occurs or is present and instances where it does not occur or is not present.

[0036] The term "population" refers to any set of subjects, including any number of subjects. A population can be selected or defined based on virtually limitless criteria. A population can be based on the number of available subjects or samples of subjects.

Alternatively, a population can be based on one or more criteria used to select the subjects in the population. Non-limiting examples of selection criteria for a population can include geographic location, race, gender, age, exposure to carcinogen(s), etc. The term "sub- population" refers to any subset of subjects within the larger population. For instance, a population can be selected based on age (e.g. subjects over the age of 50), and further subdivisions of the population can be made by defining sub-populations of subjects having disease X (e.g. diabetes) or disease Y (e.g. cardiovascular disease). Further, a sub-population, in some instances, may be determined by the response of subjects of the sub-population to a particular stimulus, wherein the response is different from other subjects of the population not within the sub-population. Experiments can determine relationships between subjects of the larger population, or between subjects of a first sub-population with or without regard to subjects of a second (or other) sub-populations of the larger population. As used herein, a population or sub-population can refer to either the set of subjects or to the samples collected from the set of subjects.

[0037] The term "donor" (or "donors") and "subjects" (or "subjects") as used herein includes any human or other animal of any age and are used interchangeably. The term includes any vertebrate, more specifically a mammal (e.g. a human, horse, cat, dog, cow, pig, sheep, goat, mouse, rabbit, rat, and guinea pig), birds, reptiles, amphibians, fish, and any other animal. The term does not denote a particular age or sex. Thus, adult and newborn subjects, whether male or female, are intended to be covered. A subject or donor is the original source of a sample.

[0038] The term "tissue" (or "tissues") as used herein includes biological cells, tissues and fluids containing cells, preferably cells, tissues and fluids containing cells that are multi- potent, or that can be used to create induced pluripotent stem cells (iPSCs), or functional cells derived from iPSCs. The cells can be obtained from large numbers of subjects as long as the cells are reprogrammed to a pluripotent state via a non-integrating method. Tissues of any kind can be collected, such as, but not limited to, amniotic fluid; cord blood or tissue;

placenta; peripheral blood cells; cells obtained from skin (including dermal fibroblasts) or hair; and cells obtained through a swab of the cheek or any alimentary passage of the donor.

[0039] The term "sub-experiment" is defined herein as: the application of a selected dose, concentration or exposure of a stimulus (e.g., chemical or biological agent) to cells or tissues originated from a single subject in the context of a specified assay, whether that application be conducted once, or in multiple replicates wherein the results of the replicates are combined in any way as to produce a single result per endpoint. For purposes of this definition, an assay may produce measurements of multiple endpoints within a single sub- experiment. Further, a stimulus in a single sub-experiment may include a combination of more than one stimulus (e.g., more than one agent, or any combination of agents).

[0040] The term "multi-dimensional array" is defined herein as: a database in any format wherein observations from measurements of the biological features of the stimulus and/or sub-experiments are recorded along with chosen descriptors of the measurement or sub- experiment, such that sub-sets of the data can be isolated from one another by subdivision by different values of one or more of such descriptors (e.g., in selecting a platform regimen and/or experiment regimen). A multi-dimensional array need not be physically arranged in dimensions. The only criterion is that the data can be separated based on distinctions in the value of a descriptor. Further, a multi-dimensional array need not be complete for all observations. Further, information in the array can be substituted, added to or subtracted from without disturbing the status of being a multi-dimensional array.

[0041] The term "platform regimen" is defined herein as: any set of variables that collectively specify a set of subjects or samples for inclusion in a population or sub- population. In some embodiments, the platform regimen determines a set of subjects or samples to a sufficient degree such that the subjects or samples are sufficiently similar that significant differences in the experimental outcomes are not attributable to biological differences among the subjects or samples themselves. While other variables related to subjects or samples may vary, the variables specified in the platform regimen do not. Thus, the platform regimen determines a population or sub-population in such a way that when, for example, two or more experiments are performed at separate times, the significant differences between results of different experiments are not due to differences among the subjects or samples themselves. The platform regimen can be defined explicitly, or implicitly as a result of the experimental actions of the researcher. Non-limiting examples of variables in a platform regimen include the donor/subject, the root cell or cell colony from which a set of cells were grown, the functional cell type of a set of cells, and an identifier of the batch or lot of a particular set of cells. Further, there may be categories and sub-categories of variables, as well as sub-divisions within such category and sub-category variables. For example, one category of variable may be "Functional Cell Types", and within that category, there might be a sub-category "Cardiomyocytes", and within that sub-category, there might be subdivisions for "Cardiomyocytes Derived Using Technology 1" and "Cardiomyocytes Derived Using Technology 2". The composition of subjects in a population defined by a platform regimen can change over time. For instance, a platform regimen defining a population of subjects which are European, over the age of 50, and have allele A of gene X may contain e.g. 10 subjects. However, as samples are obtained from increasing numbers of subjects, the same platform regimen may come to specify 25 subjects. Because the subjects of the population are determined by the selection criteria of the platform regimen, experiments performed using the initial 10 subjects are comparable to experiments performed using the latter 25 subjects (assuming other relevant variables are held constant) such that significant differences in the experimental outcomes are not attributable to biological differences among the subjects or samples themselves. The same platform regimen can be used with one or more experiment regimens in separate experiments. An Example of a platform regimen is provided in Illustration 1.

[0042] The term "experiment regimen" is defined herein as: any set of variables that collectively specify the conditions of an experiment. In some embodiments, the experiment regimen further comprises at least one stimulus to be assayed. In some embodiments, the experiment regimen determines the conditions of an experiment to a sufficient degree such that any two experiments conducted with identical experiment variables are sufficiently similar that significant differences in the experimental outcomes are not attributable to differences among the experimental conditions themselves. In embodiments comprising an experiment regimen specifying at least one stimulus to be assayed, significant differences in the experimental outcomes are not attributable to differences among the experimental conditions themselves, but may be attributable to the assayed stimulus, or stimuli. While other variables related to experimental conditions may vary, the variables specified in the experiment regimen do not. Thus, the experiment regimen determines the conditions of an experiment in such a way that when, for example, two or more experiments are performed at separate times, the significant differences between results of different experiments are not due to differences in the controlled experimental conditions themselves. In embodiments comprising an experiment regimen specifying at least one stimulus to be assayed, the experiment regimen determines the conditions of an experiment in such a way that when, for example, two or more experiments are performed at separate times, the significant differences between results of different experiments are not due to differences in the controlled experimental conditions themselves, but may be attributable to the assayed stimulus (stimuli). The experiment regimen can be defined explicitly, or implicitly as a result of the

experimental actions of the researcher. Non-limiting examples of variables of the experiment regimen include assay specification, endpoint, the type or nature of agent serving as a stimulus, and exposure specifications. Further, there may be categories and sub-categories of variables, as well as sub-divisions within such category and sub-category variables. For example, one category of variable may be "Assay", and within that category, there might be a sub-category "Endpoint", and within that sub-category, there might be sub-divisions for "Endpoint 1 " and "Endpoint 2". A non-limiting example of an experiment regimen can include conditions comprising 100 aliquots of samples, exposure to agent A, incubation at 37°C for 10 minutes, etc., and in which a stimulus to be assayed is present (e.g., agent A is assayed across the samples over a range of dosages of agent A). The same experiment regimen can be used with one or more platform regimens in separate experiments. An Example of an experiment regimen is provided in Illustration 2.

Methods for ascertaining cross-subject quantitative relationships

[0043] Recent scientific advances in the development of stem cells have theoretically opened the possibility of using these as the test platform in in vitro testing. For example, stem cells and their derivatives enable large scale production of sufficient quantities of cells from the same subject to support the multiple experiments required to test and compare multiple stimuli.

[0044] However, such cells have not been used to date for this purpose. Comparing results across subjects in large enough numbers to support conclusions at the level of large populations depends on the development of large scale cohorts of cell lines that are considered to be sufficiently biologically comparable to each other (e.g., developed under similar isolation, culturing, and differentiation conditions) to support cross-subject comparisons of the results. The development of such platforms are described in the following three published patent applications: PCT/US 14/45499, entitled "Methods for Predicting Responses to Chemical or Biologic Substances"; PCT/US 14/53819, titled "Methods for Genetically Diversified Stimulus-Response Based Gene Association Studies"; and PCT/US 15/55637, entitled "Methods for Conducting Stimulus-Response Studies with Induced Pluripotent Stem Cells Derived from Perinatal Cells or Tissues," each of which are incorporated by reference herein in their entireties.

[0045] The present methods build on that capability by specifying the methods, designs and subsequent processes to examine whether responses to a stimulus do indeed correlate across subjects within such populations, including when both incidence and severity of response are considered.

[0046] Specifically, the methods provided herein assist in addressing issues of the following description: "When analyzed across the subjects of an experimental population, without aggregating the results of subjects into a collective measure of the population (e.g., average, etc.), what is the quantitative relationship among two or more designated measures of biological activity or features ("response indicators"; e.g., the scores on two endpoints of assays, or the scores of an assay contrasted with a biological feature such as size), referred to here as Measure A and Measure B? Such relationships can be:

[0047] Case 1 : Measure A and Measure B each exhibit a narrow range of values across the subjects of a population, but the value of Measure A in any subject is unrelated to the value of Measure B in that same subject;

[0048] Case 2: Either Measure A or Measure B (but not both) exhibits a wide range of values across subjects of a population, but the value of the other Measure in a subject is unrelated to the value of the wide-ranging Measure; [0049] Case 3 : Both Measure A and Measure B exhibit wide ranges of values across subjects of a population, but the value of Measure A in an subject is unrelated to the value of Measure B in that same subject;

[0050] Case 4: The value of Measure A in a subject bears a mathematical relationship to the value of Measure B in that same subject. In such cases, the relationship may exist within a range of error. For example, the case may include a situation wherein the value of Measure A equals the value of Measure B times two, plus or minus 20 percent. This relationship can be of any mathematical form, including but not limited to: linear or non-linear correlations, step functions (e.g., a particular constant value of Measure A when Measure B lies within a specified range, but a different constant value of Measure A when Measure B lies within another range), etc. ;

[0051] Case 5 : Any combination of the above cases, such as when Case 4 exists for certain ranges of Measure A, but Case 1 exists in other ranges of Measure A.

[0052] The methods provided herein are useful for ascertaining cross-subject quantitative relationships between response indicators among multiple subjects in a population using multipotent or pluripotent stem cells. Use of multipotent or pluripotent stem cells provides an infinite pool of samples from a subject, permitting experimentation to occur over broad ranges of time and for a wide array of purposes. Further, multipotent or pluripotent stem cells provide starting materials from which essentially all cell types of the same subject can be derived. Thus, use of multipotent or pluripotent stem cells permits experimentation essentially on any of a subject's cell types and at essentially any time. While this greatly expands the experimental potential of researchers, issues remain regarding comparison of data obtained from separate experiments or trials, or regarding defining all the population and experimental variables before beginning an experiment. By defining in a platform regimen the population of subjects to be analyzed, and by defining in an experiment regimen the experimental controls and stimulus to be assayed, separate experiments using multipotent or pluripotent stem cells are comparable. Such comparisons between experiments performed separately can be made with a very high level of confidence that differences observed are not due to either the controlled biological differences of the subjects or samples themselves, or to the controlled experimental conditions. Thus, the methods herein provide for a means to ascertain cross-subject quantitative relationships between response indicators among multiple test subjects in a population, wherein the populations and conditions can be defined either before or after samples are obtained from subjects, and before or after experiments are performed on samples. As a result, subject or sample controls and experimental controls can be determined a priori or post-hoc in relation to collecting samples or performing wet-lab experiments. The ability to analyze pre-existing data, newly obtained data, or mixtures of pre-existing data and newly obtained data, to determine relationships between response indicators using an infinite source of cells and cell types (derived from multipotent or pluripotent stem cells), permits innumerable, scientifically-controlled comparisons. For example, adherence to platform regimens and experimental regimens enables a researcher to compare the results of two experiments conducted on a population of 10 or more subjects on an individual subject by individual subject basis (rather than on the basis of aggregated statistics) even if one of those experiments had not been conceived of at the time the other experiment was conducted. Further, should the results of the second experiment in a pair of experiments produce an insight that suggests that a third experiment should be developed (and results subsequently compared to the results of the first or second experiment), this too can be made feasible by the proper use of such regimens. In the absence of platform and experiment regimens, the researcher lacks a framework for establishing which results of many possible experiments can be scientifically compared versus which cannot, (due to lack of control of key variables other than the one being studied).

[0053] In some embodiments, methods for determining a relationship between two or more response indicators in a population comprise a) selecting a platform regimen comprising a set of variables that selects separate samples from a population of at least 10 subjects, wherein the samples comprise proliferative multipotent or pluripotent stem cells; b) selecting an experiment regimen comprising a set of controlled experimental conditions and at least one stimulus, wherein the at least one stimulus is applied to or combined with the sample to generate a response; c) assaying the samples for response to the at least one stimulus, wherein a response indicator comprises a quantified response endpoint to the at least one stimulus; d) determining for each sample an intra-subject relationship between two or more response indicators; and e) comparing in a cross-subject study the intra-subject relationships between the two or more response indicators to determine a cross-subject relationship between the two or more response indicators.

[0054] A platform regimen comprising a set of variables can be selected, wherein the set of variables (and hence, the platform regimen) selects a population of subjects or samples from subjects. A platform regimen specifies a set of variables used to select a population used to evaluate one or more response indicators. In some embodiments, the platform regimen selects at least about 10, 15, 20, 25, 30, 35, 40, 45, 50 or more samples to be included in a method. As used herein, terms directly relating to selection of the set of variables in the platform regimen are used interchangeably with terms directly relating to selecting a platform regimen. Selecting a platform regimen can be performed prior to, during, or after collecting samples and/or prior to, during, or after performing experiments (e.g. prior to, during, or after assaying samples for a response). The set of variables selected depends on the variables which may influence conclusions about the response indicator. As a non-limiting example, a set of variables selected in a platform regimen used to evaluate a response indicator to exposure to an agent may comprise particular ages and races of subjects, but exclude height and weight of a subject. Such a selected platform regimen would be appropriate in situations where it is believed or hypothesized that age or race may affect a subject's response to the agent, but where a subject's height and weight would not influence a subject's response to the agent. Thus, in a first experiment, the proper biological features and other variables related to selection of the population which may affect responses are controlled for, such that differences in results observed are not due to the controlled variables of the population. In this manner, two or more separate experiments may also be compared (where the experiment regimens are also identical) because the platform regimen controls for the biological features and other variables related to selection of the population which may affect responses.

[0055] An experiment regimen can be selected, wherein the experiment regimen comprises a set of controlled experimental conditions and at least one stimulus, wherein the at least one stimulus is applied to or combined with the sample to generate a response. The conditions to be controlled and the stimulus to be applied or combined define the

experimental conditions apart from the subjects included. An experiment regimen defines (a priori or post-hoc) a set of conditions used to conduct an experiment which evaluates a subject's response to a condition or stimulus. Selecting an experiment regimen can be performed prior to, during, or after collecting samples and/or prior to, during, or after performing experiments. Like the platform regimen, the set of conditions selected for use in an experiment regimen depend on the experimental conditions which may influence conclusions about the response indicator. As a non-limiting example, a set of experimental conditions selected in an experiment regimen used to evaluate a response indicator to exposure to an agent may comprise a range of dosages of the agent, but exclude temperature controls. Such an experiment regimen would be appropriate in situations where it is believed or hypothesized that varying dosages of an agent may affect a subject's response to the agent, but where the temperature at which a subject's sample is exposed to an agent would not influence a subject's response to the agent. Such an experiment regimen can further specify at least one stimulus to be assayed (e.g., varying dosages of an agent, varying amounts of physical stressors such as pressure, etc.).

[0056] Although both regimens control for variables that may influence experimental outcomes, the platform regimen selects subjects or samples used in experiments whereas the experiment regimen selects the conditions for an experiment. The platform regimen and experiment regimen need not control for every variable; they need only control for those biological variables of the population and those experimental conditions which may influence experimental outcomes such that significantly different results obtained are not due to those biological variables of the population and experimental conditions which may influence experimental outcomes.

[0057] Selection of an experiment regimen can be a separate consideration from selection of a platform regimen, although a method user need not recognize the distinction to perform a herein disclosed method. For many variables, categorization will be quite obvious.

However, in some instances, the distinction as to whether a variable constitutes a variable of the platform regimen or experiment regimen can be blurred. As an example, growing and expanding cells in a first reagent or in a second reagent can affect cellular function, structure, or morphology. The platform regimen could be defined as comprising samples from subjects over age 50 which are differentiated into cardiomyocytes grown and expanded in a first reagent or a second reagent. Alternatively, the experiment regimen could be defined as comprising first growing and expanding cells in the presence of a first reagent or a second reagent, and exposing the cells to a range of 1 -100 μg/mL of an agent for 10 minutes. Thus, categorization of variables is of lower importance compared to whether the selection of the population and the experimental conditions are adequately controlled between the two regimens, such that significantly different results obtained are not due to those biological variables of the population and experimental conditions which may influence experimental outcomes. [0058] Collectively, the platform regimen and the experiment regimen specify the population on which experiments can be performed, and the conditions of the experiment(s) to be (or already having been) performed. An advantage of the platform regimen and the experiment regimen is that either or both can be specified before or after a wet-lab experiment is performed. For instance, the proper platform regimen can be specified first, such that the proper subjects are first identified and then an experiment is performed.

Alternatively, a data set derived from a large number of samples subjected to different experiments can be mined for the proper subjects specified by the platform regimen.

Additionally, the proper experiment regimen can be specified first, such that the proper experimental conditions are first identified and then an experiment is performed.

Alternatively, a data set derived from a large number of samples subjected to different experiments under varying conditions can be mined for the proper data sets derived from the proper experimental conditions specified by the experiment regimen. Further still, newly obtained data sets from new experiments can be compared to old data sets from prior experiments. This greatly expands the universe of comparable data sets while maintaining a high level of confidence that significantly different results observed are not due to the controlled variables of the platform regimen or experiment regimen.

[0059] As used herein, a sample from a subject includes any biological material which contains viable cells of the subject, e.g., blood, skin, bone, sputum, hair, nails, mucosal and other bodily secretions, etc. In some embodiments, a sample can be any purified set of cells from a mixture of biological materials, e.g. keratinocytes obtained from a skin biopsy, hematopoietic stem cells obtained from a bone marrow biopsy, etc. In some embodiments, a sample can comprise proliferative multipotent or pluripotent stem cells. For example, a sample can comprise induced pluripotent stem cells (iPSCs) derived from biological material which contains viable cells. A sample may be obtained in one form from a subject (e.g. as blood in a blood draw), and converted to and stored as another form (e.g. iPSCs in a cryofreezer bank). In some embodiments, a sample is converted into iPSCs (still referred to as a sample of the subject) using a stem cell differentiation kit. In some embodiments, the samples can be non-embryonic stem cells or derived from non-embryonic stem cells.

[0060] One advantage of converting samples into iPSCs is the creation of an infinite source of samples from a subject. The iPSCs can be grown and expanded, even reiteratively, to maintain the infinite source of samples. Once a sample is collected from a subject and converted to iPSCs, a portion of the iPSCs need only be grown and expanded to replenish the iPSC supply when the supply runs low. In some embodiments, iPSCs are grown and expanded to first generate a large supply prior to assaying the samples. Another advantage is that iPSCs and other multipotent or pluripotent stem cells can be converted into essentially any cell type. A single sample of biological material (e.g., blood) from a subject can be converted into iPSCs, from which essentially any cell type of interest (e.g., cardiomyocytes, neuronal cells, etc.) can be derived. Conversion of disparate samples to a generic, infinite source of essentially all cell types permits the cross-subject comparison of different samples obtained from different subjects at different times.

[0061] The conditions for growth and expansion of iPSCs can affect structural, functional, and/or behavioral aspects of the sample. Stem cells such as iPSCs may require well-controlled growth and expansion conditions to generate accuracy, precision, and repeatability in downstream experiments. In some embodiments, the herein described methods determine relationships between response indicators related to growth and expansion conditions. As a non-limiting example, the methods can determine relationships between response indicators of different reagents used in the growth and expansion medium. Cells can be grown and expanded in a first medium comprising a first agent and separately in a second medium comprising a second agent, wherein the first agent and the second agent are reagents of the growth and expansion medium. Traditionally, a researcher may analyze whether one or only a few cell lines are affected by changes in the growth and expansion medium. The disclosed methods can determine the relationship between changes in the growth and expansion medium (e.g., between the first and second agents on growth and expansion of the stem cells) across samples from numerous (e.g., at least 10) subjects. The cross-subject study can determine whether changes in the growth and expansion medium affect structural, functional, and/or behavioral aspects of none or all samples or, alternatively, only a sub-population of the samples.

[0062] A similar concept applies to methods to ensure quality control in reiteratively produced stocks of iPSCs. In some embodiments, the samples assayed comprise a parental aliquot of each sample before growth and expansion, and an offspring aliquot of each sample obtained after growth and expansion. Thus, relationships between any structural, functional, and/or behavioral changes in cells produced after growth and expansion (offspring) as compared to the original iPSC stock (parent) can easily be determined. [0063] Multipotent or pluripotent stem cells such as iPSCs can be differentiated into essentially any cell type, using methods and/or kits known in the art. This feature of iPSCs allows for the production and experimentation of essentially any cell type of a subject without having to collect all cell types from the subject (of which the practical considerations and invasiveness are extensive). In some embodiments, stem cells are differentiated into at least 2, 3, 4, 5, or more different functional cell types. The number or range of cell types from which stem cells may be differentiated into is limited only by the ability to differentiate stem cells to a particular cell type. The different functional cell types of each sample can be assayed and compared. In such an example, the methods can determine a relationship between the response indictor of a first functional cell type and the same (or another) response indicator of a second functional cell type. Although each functional cell type of one subject theoretically contain identical genomes, the expression of genetic information may vary between functional cell types. Thus, within the same subject, relationships between response indicators may vary between functional cell types of that subject depending on, for instance, variable gene expression or function in different functional cell types. In some embodiments, the methods can determine an intra-subject relationship comprising a relationship between differential gene expression or function in two or more functional cell types. In some embodiments, differential gene expression or function may be related to two or more alleles of a gene.

[0064] A sample can be obtained from a subject by any method known in the art useful to obtain the particular sample type. For example, a sample can be obtained by blood and/or plasma draw (e.g. venipuncture), blood donation and other phlebotomy techniques, puncture methods such as lumbar puncture or arthrocentesis, swab, biopsy, passive drool, placement of a sample into a container by the subject as in e.g. semen or breast milk donation, biopsy and/or surgical excision, forensic collection of biological material left by a subject, and other methods.

[0065] A response indicator (or indicator of a response) includes any variable, quality, condition, or characteristic which provides information about a sample's (and hence, a subject's) response to a particular stimulus or condition. The term is intended to broadly encompass essentially any variable, quality, condition, or characteristic which can exist in or result in at least two or more states, values, or inputs, and wherein at least one state, value, or input provides information about the subject's response to a condition or stimulus. A response indicator can, but need not, provide information about one, multiple, or all subjects of a population. In some embodiments, a response indicator may provide conclusive information about the biological response of subjects of a sub-population, but provide inconclusive information for the remaining subjects of the broader population not within the sub-population.

[0066] A response indicator can be quantitatively and/or qualitatively evaluated.

Although not required, a relationship between two or more response indicators can be better articulated when at least one of the response indicators is quantified. Thus, in some embodiments, a first response indicator comprises a quantified response endpoint to at least one stimulus. Typically, a measured, quantified response endpoint is obtained in an assay of the stimulus. As a non-limiting example, the assayed stimulus can be a range of dosages of an agent, and the response indicator may be a particular dose at which a pre-defined percentage of cells do not survive (e.g. from toxicity of the agent). In such an example, the percentage of cell death (derived from the quantified endpoint of the assay) indicates the response to varying doses of the agent. In some embodiments, a response indicator can comprise a qualitative evaluation of a response. As a non-limiting example, the assayed stimulus can be a range of dosages of an agent, and the response indicator may be morphological changes in cells exposed to the agent. In such an example, the qualitative observations (e.g., cellular elongation, membrane perturbations observed by microscopy, general nuclear swelling, etc.) indicate the response to varying doses of the agent. In other embodiments, numerous response indicators can comprise quantifiable endpoints. Thus, in some embodiments, samples can be assayed with a first stimulus comprising a first agent and a second stimulus comprising a second agent, wherein the first response indicator comprises a quantified response endpoint to the first agent and a second response indicator comprises a quantified response endpoint to the second agent. In some embodiments, two or more response indicators can be from different assays, an assay and a non-assay experiment, or can be for the same assay (e.g., different endpoints of the same assay).

[0067] In providing information about a sample's response to a stimulus, a response indicator can provide information of a different nature about the sample, from which a relationship between two or more response indicators can be determined. In some embodiments, a response indicator can provide structural, functional, and/or behavioral information about a sample. In some embodiments, assaying the samples with a first stimulus provides information of one type about the samples, and assaying the samples with a second stimulus provides information of another type about the samples. For instance, in some embodiments, assaying the samples with the first agent provides structural information about the samples, and assaying the samples with the second agent provides functional or behavioral information about the samples. In some embodiments, two or more response indicators may result from the same stimulus, wherein at least two of the response indicators provide information of a different type. As a non-limiting example, the assayed stimulus can be a range of dosages of an agent, the first response indicator can be the measured production of a metabolite, and the second response indicator can be the visual observation of cell membrane damage. Thus, the methods can determine a relationship between the production of a metabolite (functional information) and cell membrane damage (structural information) in response to dosages of an agent.

[0068] The herein disclosed methods determine relationships between two or more response indicators. Using the example response indicators in the preceding paragraph, the methods determine the intra-subject relationship between production of a metabolite and the morphological change (e.g., cell elongation) in response to varying dosages of an agent. An intra-subject relationship is a relationship between two or more response indicators in a single subject or single sample defined by the platform regimen. Non-limiting examples of ways to articulate an intra-subject relationship between two or more response indicators include direct, inverse, multiplicative, mathematical, qualitative (e.g., larger vs. smaller), etc.

[0069] Data obtained for a single subject (e.g., in a sub-experiment) can be compared to data obtained from other subjects in a cross-subject study. For example, the relationship between two or more response indicators in a single subject can be evaluated across a population of subjects (e.g., at least 10 subjects). Thus, a cross-subject study is the comparison of data obtained from multiple intra-subject experiments (sub-experiments) across the members of the cross-subject study. For instance, an intra-subject sub-experiment may identify a single subject wherein no correlation exists between the two or more response indicators, but a cross-subject study may identify sub-populations of subjects in which a correlation does exist between the two or more response indicators.

[0070] Cross-subject studies can identify subpopulations of interest in which, for example, the relationship between two or more response indicators is unique. As an example, in some embodiments, the methods identify a sub-population of subjects for which a particular agent provides increased efficacy, reduced side-effects, or both compared to other subjects of the population, which are more responsive to a different agent. Once such a subpopulation is identified, further comparisons can be made to determine the nature, cause, or extent of the relationships between sub-populations. For instance, the genetic profile of at least one subject in the sub-population can be compared with the genetic profile of at least one subject of the population that is not within the sub-population, (e.g., a subject from a separate, non-overlapping sub-population). Such a comparison of genetic profiles may potentially identify a genetic determinant(s) correlated with the sub-population's response to the agent providing increased efficacy, reduced side-effects, or both.

[0071] As would be understood by one of skill in the art, the number, type, and nature of response indicators are virtually limitless, and not all will be relevant to each experimental situation. Thus, the set of response indicators which provide information about a particular subject's response to a condition or stimulus can be virtually infinite, limited only by changes which can be observed. Considering the example in which samples from a population of subjects are assayed with varying concentrations of an agent, the response indicators can comprise, for example, percent of cell death, quantification of cell growth, amount of metabolite production, gene induction/repression, percent or rate of cell differentiation, enzyme activation, substrate modification, or virtually limitless other observations. In some embodiments, the methods identify and characterize response indicators. In some embodiments, the methods identify and characterize relationships between two or more response indicators in a population.

[0072] In some embodiments, the experiment regimen specifies at least one stimulus. In some embodiments, the at least one stimulus is applied to or combined with the samples to generate a response. In some embodiments, the samples can be assayed with the at least one stimulus. In some embodiments, the samples can be assayed with two or more stimuli. The stimuli can be related or unrelated. For instance, two or more stimuli can comprise two closely related isomers of a particular drug, or alternatively, can comprise exposure to a drug and exposure to radiation.

[0073] The stimulus is not particularly limited, but should be one in which an observation can be made when samples are exposed to or assayed with the stimulus. The stimulus can comprise any agent, radiation, or physical stressor. In embodiments in which the stimulus comprises an agent, the agent can comprise any biological or chemical agent which can be exposed to a biological sample collected from a subject. Numerous agents may be of interest and hence, are not particularly limited. Non-limiting examples of agents include biological agents such as antibodies, proteins, lipids and glycolipids, steroids, hormones,

neurotransmitters, viruses, viral vectors, bacteria, liposomes, biological extracts such as plant extracts, and chemical agents such as small molecules, carbon-based molecules, synthetic and derivative molecules, drugs such as therapeutic drugs, and a wide range of other agents. For example, certain therapeutic and/or pharmaceutical compounds may be of interest to a researcher for a particular situation, but a different set of therapeutic and/or pharmaceutical compounds may be of interest in a different situation. As an example, in some situations in which it is desirable to investigate which agents are effective to treat cancer, the agent can be an FDA-approved or experimental drug administrable to treat cancer. Alternatively, in some situations in which it is desirable to investigate which agents are effective to treat a condition but avoid cardiotoxicity, the agent can be an FDA-approved or experimental drug administrable to treat that condition that is evaluated for cardiotoxic effects in the disclosed methods. Thus, the nature of the experiments and the questions being investigated in part determine selection of a particular stimulus. In some embodiments, the stimulus comprises radiation. Non-limiting examples of radiation stimuli comprise ionizing radiation including, but not limited to, x-ray radiation, γ-radiation, energetic electron radiation, ultraviolet radiation, thermal radiation, cosmic radiation, electromagnetic radiation, nuclear radiation, or any combination thereof. In some embodiments, the stimulus comprises physical stressors. Non-limiting examples of physical stressors comprise pressure, temperature, agitation or shaking, etc.

[0074] Samples can be assayed with one or more stimuli under a wide array of conditions. Some conditions which can, but not necessarily need to, influence observable responses of a sample to exposure to a stimulus include duration of exposure, incubation temperature, agent concentration, amount of sample, agent activation state, presence of additional factors (e.g. co-factors, substrates, enzymes, etc.), condition of samples (e.g. clumped vs. dispersed cells), and other variables. The experiment regimen controls for the variables of interest which can influence observations made when samples are assayed with a stimulus. Preferably, the samples are assayed with the stimulus for a time sufficient for a response to be observed and recorded.

[0075] Samples can be exposed to or assayed with a stimulus in numerous ways known to those of skill in the art. Assays can be designed to control for a particular condition, e.g. agent concentration. In some embodiments, the agent is assayed with replicates of numerous samples in a population study. More than one stimulus (e.g., agent) can be included in an assay. As an example, samples can be assayed with a constant concentration of a first agent and increasing (or decreasing) concentrations of a second agent. Additional variables can be simultaneously tested in the same assay. For instance, samples assayed with a constant concentration of a first agent and increasing (or decreasing) concentrations of a second agent can additionally be incubated at varying temperatures. Further, the methods can compare samples from numerous subpopulations. Thus, two or more subpopulations can be assayed with an agent(s) under one or more conditions.

[0076] As used herein, the terms "exposing", "assaying", or other forms referencing the combining of any two or more components (e.g. an agent and a sample) include numerous arrangements in which the components are exposed, applied to, combined and/or assayed. A sample from a subject, for instance, can be subdivided into two or more portions, and further divided into even smaller sub-portions. For instance, a sample from a subject can be first divided into a number of aliquots to be stored in a freezer for future use. Any one aliquot may be further subdivided into separate wells of a microtiter plate in a manner substantially identical between all subdivisions of the sample. Samples may be differentiated into different functional cell types, which are also, in some instances, still referred to as samples of the subject. For simplicity, any given portion, sub-portion, or differentiated portion of a sample is referred to herein simply as a sample of the subject, unless clearly and unambiguously stated otherwise. For example, a sample can be assayed with two agents in various ways. In one instance, the exposure of more than one agent to a sample can include the exposure of two agents (Agent 1 and Agent 2) to the exact same portion of a sample (e.g. both Agent 1 and Agent 2 are added to an aliquot of sample 1 in well Al of microtiter plate). In another instance, the exposure of more than one agent to a sample can include the separate exposure of two agents (Agent 1 and Agent 2) to separate portions of a sample (e.g. Agent 1 is exposed to an aliquot of sample 1 in well Al of a microtiter plate but is not added to well A2, while Agent 2 is exposed to a separate aliquot of sample 1 in well A2 of a microtiter plate but is not added to well Al). The above examples are not intended to be limiting in any way, but rather demonstrative of the various uses of the terms exposing, assaying, or other forms referencing the combining of any two or more components, particularly samples.

[0077] A user may choose to base conclusions as to the relationship, or differences, between the populations or sub-populations on any statistical, mathematical or graphical comparison of the results of the experiments. Statistical analysis of the data also depends on the nature of the study performed and is inclusive of a wide array of statistical studies, programs, and techniques known to those of skill in the art. Whether a difference is significant, or statistically significant, depends on the parameters selected for the method. In some embodiments, the threshold for statistical significance is determined by a statistical p- value. As non-limiting examples, any or all of the following could be used: comparisons of each population or sub-population's mean mode or medians; comparisons of any percentile of results (e.g. the 80th percentile) of the compared population or sub-populations; a notation that 44 percent of the observations for levels of effects on a first sub-population fall substantially below the lowest observed level of effect for any subject of a second sub- population, etc. This list is illustrative, and is not intended to be comprehensive. Any applicable technique to evaluate the difference in responses of sub-populations to exposure to an agent may be used, as would be understood by one of skill in the art. Guidance for evaluating the significance of any differences in data can be found, for example, in Navidi, W.C. et al, Elementary Statistics, McGraw-Hill Higher Education (Oct. 2014); and in Lane, D. et al , Online Statistics Education: A Multimedia Course of Study (available online at onlinestatbook.com), each of which are fully incorporated by reference herein.

[0078] Methods for the use of stem cells (and/or cells derived from stem cells) that have been reprogrammed and/or expanded from cells taken from representative (broadly defined) subjects of a population for the purpose of ascertaining patterns of behavioral or structural responses of those cells to stimuli are described herein.

[0079] In some embodiments, the methods comprise determining, in at least 10 subjects, the cross-individual quantitative relationship between/among two or more response indicators to the presence of a stimulus, wherein (1) at least one of the indicators is a quantitative measurement of a specific endpoint resulting from an assay; (2) a quantitative relationship between the indicators is established at the level of an individual subject (referred to herein as an "intra-subject relationship"), and the cross-subject quantitative relationship being determined compares the intra-subject relationships across multiple subjects of the population; and (3) an application of the method includes the following steps, undertaken in any appropriate order: a) providing samples comprising proliferative multipotent or pluripotent cells derived from at least 10 test subjects, wherein the samples are maintained separately for each subject; b) Growing or expanding the samples as necessary to obtain sufficient quantities of cells to support multiple experiments as described below, and optionally, differentiating and/or aggregating the cells as required to support those experiments; c) Formally or informally establishing (a priori, or post-hoc) a platform regimen of attributes (which may include but need not be not limited to: specifications of the samples from which the platform regimen was established, specifications of differentiated and/or aggregated cells, and cell batch identities), wherein the particular specification of an element of a variable in the platform regimen serves as the default value for that element which cannot be deviated from during any experiment under this application of the method, except wherein the deviation from the element is intended to be a source of variation that results in an indicator being distinguished from other indicators such that they can be compared to form an intra-subject relationship; d) Formally or informally establishing (a priori, or post-hoc) an experiment regimen (which may include but is not limited to: assay specifications, endpoint specifications, agent specifications, exposure specifications, etc.), wherein the particular specification of an element of a variable in the experiment regimen serves as the default value for the element which cannot be deviated from during any experiment under this application of the method except wherein the deviation from the element is intended to be a source of variation that results in an indicator being distinguished from other indicators such that they can be compared to form an intra- subject relationship; e) Conducting two or more experiments wherein: i. The experiment comprises assays conducted on the cells described in (b) for at least 10 test subjects from the population; and ii. The experiment is conducted in such manner that each element of each variable in the platform regimen and experiment regimen is adhered to, except for the chosen elements intended to be a source of variation that results in an indicator being distinguished from other indicators such that they can be compared to form an intra-subject relationship; f) Creating data sets for each subject on which the experiments in Step e) above were successfully conducted, wherein the data sets include, but are not limited to, quantitative results for the specified endpoints from two or more of the experiments; and g) Mathematically analyzing the data sets, using any statistical technique that compares or relates the data sets while preserving the data at the individual subject level (including, but not limited to linear regression, non-linear regression, clustering, threshold analyses, and step function techniques).

[0080] In some embodiments, the method comprises the following steps, undertaken in any order: a) A comparison of biological features and reactions to be investigated is determined, and the multi-dimensional array of data used to conduct such

investigations is defined. For example, this data may include: the number of subjects to serve as test subjects; for each subject, the types of cells on which assays are to be performed; designation of the biological feature measurements and assays to be performed, including the endpoints on which data is to be collected for later comparisons; and designation of the various stimuli to be applied in the assays; b) A population for which inferences are to be drawn from the experiments and analyses is specified, and the criteria for selecting the representative population is specified, and specific subjects who constitute the population are identified; c) Source tissues are collected from each subject; d) For each test subject, source tissues are converted to cell colonies that will serve as the "root source" of all cells used in subsequent measurements and assays. Preferably, a single clonal cell colony will exist for each subject. More preferably, the single clonal cell colony for each subject will contain an iPSC colony; e) If required by the particular measurements or assays chosen for any sub- experiment, cells from the cell colony for each subject are differentiated into the required cell type; f) Measurements and assays are conducted, and data from sub-experiments recorded in a multi-dimensional array as outlined in step (a); g) Specific comparisons to be examined across the results from subj ects are defined, and cross-subject statistical relationships to be measured are selected, and the defined analysis is conducted; and h) Step (g) is repeated as necessary.

[0081] Numerous embodiments of the present disclosure are envisioned. In a first embodiment, the multiple stimuli, for whom the results are being compared, are different agents, or are different exposure/dose concentrations of the same agent.

[0082] In another embodiment, the reactions being compared (in terms of incidence or magnitude), within a single subject, occur in the same cells or assay; within different cells or assays, but wherein the reactions being compared are within the same cell types, or are being compared across different cell types.

[0083] In another embodiment, the mathematical transforms that relate the numerical measure of biological features or one endpoint in one reaction (e.g., response) to another endpoint in the same reaction within the same subject are additive, multiplicative, or denoted by any mathematical algorithm.

[0084] In another embodiment, the mathematical transforms that relate biological features or reactions (e.g., response) within a single subject are found to bear any particular relationship to the transforms of any other subject, including but not limited to imperfect correlations.

[0085] Disclosed herein are methods which can be used with an array of materials useful for carrying out any one or more disclosed method. Where a method is disclosed and a number of modifications to the method are discussed, each and every combination and permutation of the method, and the modifications that are possible, are specifically contemplated unless specifically indicated to the contrary. Likewise, any subset or combination of these is also specifically contemplated and disclosed. This concept applies to all aspects of this disclosure. Thus, if there are a variety of additional steps that can be performed, it is understood that each additional step can be performed with any specific method step or combination of method steps of the disclosed methods, and in any order or permutation, unless otherwise indicated, and that each such combination or subset of combinations is specifically contemplated.

[0086] Publications cited herein are hereby specifically incorporated by reference in their entireties and at least for the material for which they are cited.

EXAMPLES

[0087] Some exemplary embodiments of the present invention will now be illustrated in the following specific, non-limiting examples. The examples below are intended to further illustrate certain aspects of the methods and compositions described herein, and are not intended to limit the scope of the claims.

Example 1

[0088] Two agents, such as pharmaceutical compositions, are directed to the same therapeutic goal. The first agent, which is already on the market, is deemed to be effective, when administered to humans, in 35 percent of cases. The second agent appears to be effective in about the same portion of cases. In addition, the two agents appear to have very similar overall incidence of adverse side effects. In the absence of additional information, there appears to be little incentive for regulators to approve the marketing of the second agent, as there would appear to be no "net" benefit to society, while there always exists a possibility that any newly approved drug will later prove to exhibit a previously undetected toxicity.

[0089] However, society could benefit if patients who cannot tolerate the first agent (due to e.g., adverse side effects) are able to tolerate the second agent.

[0090] Under current practices, there is no practical way to determine whether some patients who have an adverse reaction to the first agent do not have an adverse reaction to the second agent. Human in vivo testing of both agents on the same subjects would not be allowed on both scientific and ethical grounds, and the presence of residual amounts of either agent would compromise any findings as to the effects of the alternative agent tested.

Separately, an attempt to establish the partial or non-overlap of impacts via statistical comparisons of separate patient populations for the two agents would face serious scientific challenges. Alternatively, animal testing would be irrelevant, as the differential impacts of the variations of genetics among human beings is the very factor to be investigated. Finally, current in vitro testing methodologies, focused as they are on a single representative human cell donor, also fail to provide the necessary human genetic diversity.

[0091] A researcher establishes that a particular in vitro assay measures the root impact on cells that accounts for some of the adverse side effects of the two agents. The researcher then conducts the assay on stem cells derived from a large population of subjects, using the first agent as the stimulus, administered at a specified dose concentration of interest (i.e., one corresponding to the therapeutic dose). The researcher then quantitatively measures the degree of impact (as determined by the assay) on each of the subjects separately, recording their individual scores.

[0092] The researcher then repeats the assay on identical cells taken from the same population of subjects, but this time under stimulus by the second agent at its dose concentration of interest (again, the therapeutic dose, which may or may not be the same does as in the case of the first agent), and again quantitatively measures the degree of impact on each subject separately, recording the scores.

[0093] Finally, the researcher runs a set of statistical tests, comparing the results across all of the subjects, of the cross-subject relationship between the effects of the two agents on each subject individually (e.g., intra-subject relationship), including linear and non-linear correlation models. These statistical tests determine whether there is a statistically relevant difference in the tested side effects of the two agents (either measured in incidence of impact above a threshold limit, or in the relative degree of impact on the subjects in the population) when results are measured at the individual subject level.

[0094] The above procedures are repeated across a spectrum of in vitro assays that measure potential adverse side effects of different types.

[0095] From this work, the researcher is able to establish that, while there is some overlap in the subjects of the population on whom the two agents produce adverse side effects, there are also significant numbers of subjects who are vulnerable to adverse effects from the first agent but not the second agent. Thus, there is a potential benefit to society to marketing the second agent and a case to be made for regulatory approval.

Example 2

[0096] During drug development, many agents, such as pharmaceutical compositions, fail due to clinical trial patients exhibiting various forms of cardiotoxicity, wherein cardiotoxicity is manifested in the form of irregular behavior of the heart. These effects include

acceleration or lengthening of the heartbeat, and arrhythmia, among others.

[0097] Each of these behavioral manifestations may be associated, or even caused by, other impacts of the agent that are structural in nature, such as damage to cell membranes, nuclei, or mitochondria. However, until now, it has been difficult to link degrees of structural damage to degrees of behavioral shifts, for a number of reasons. First among these is the fact that detecting such structural changes requires direct access to the tissues themselves as well as destructive testing - both of which are ethically and scientifically infeasible in vivo.

[0098] A researcher is interested in studying potential linkages between structural damage to cardiomyocytes and behavioral aberrations when the cardiomyocytes are challenged by several agents of interest. The researcher obtains cellular tissue from about 10 or more (e.g., 30) subjects representing a cross-section of the population of interest. The researcher then converts the tissues of each subject into iPSCs using a stem cell

differentiation kit (commercially available from Reprocell-Stemgent, Lexington, MA), and follows the protocols and standards described in one or more of the following published patent applications: PCT/US 14/45499, entitled "Methods for Predicting Responses to Chemical or Biologic Substances"; PCT/US14/53819, titled "Methods for Genetically Diversified Stimulus-Response Based Gene Association Studies"; and PCT/US15/55637, entitled "Methods for Conducting Stimulus-Response Studies with Induced Pluripotent Stem Cells Derived from Perinatal Cells or Tissues". This process further involves the selection of iPSC colonies that have been produced from a single iPSC cell, and are thus clonal colonies.

[0099] After the iPSCs have expanded into a large number of cloned iPSCs, the researcher then differentiates the cells into cardiomyocytes, using the protocol and technology described in US Patent No. 7,029,913, assigned to the Wisconsin Alumni Research Foundation ("WARF").

[00100] As a result, the researcher is able to conduct two sets of assays on clonal colonies from each of the subjects in the cross-section of the population, while being confident that the cardiomyocytes for any one subject are scientifically identical between the assays.

[00101] One assay, the GE Cell Health Assay, (commercially available from GE Health Care Life Sciences, Marlborough, MA) produces quantitative results for structural endpoints: cell membrane integrity, nuclear integrity, and mitochondrial health at a time period 24 hours after the administration of an agent to each cardiomyocyte. The second assay, based on a calcium dye that fluoresces upon electrical activity of the cardiomyocytes (commercially available from Molecular Devices, Palo Alto, CA), is also conducted 24 hours after the administration of the same agent, and produces data on the following heartbeat

characteristics: heart beat rate and arrhythmia.

[00102] The researcher combines the data from the two assays for each subject, and across the population of subjects, to form a complete data set for the cross-section of the population, for each agent. By running linear and non-linear regressions of various heartbeat behavioral characteristics (as the dependent variables) against the various structural factors (as the independent variables), the researcher is able to establish whether there appear to be statistically valid linkages across the population between the structural factors and the behavioral characteristics for each agent.

[00103] This information enables the researcher to focus efforts on understanding any causality of structural changes on behavioral characteristics to only those relationships that have statistical significance.

Example 3

[00104] During development of an investigational new drug, researchers conduct an assay on the investigational new drug, or agent, using cardiomyocytes derived from iPSCs taken from a population of 50 subjects. The researchers find that the agent (referred to herein as CI) produces various levels of adverse impact (as measured by a particular endpoint of the assay) on certain subjects in the population.

[00105] As CI proceeds through the drug development process, the researchers alter the chemistry of the agent in an effort to improve the solubility of the agent. The altered agent is referred to herein as C2.

[00106] At a later point in the process, the researchers desire to know whether the chemistry changes between CI and C2 have mitigated or exacerbated the levels of adverse impact. In addition to measuring how C2 compares to CI on aggregate measures (such as the percentage of the population whose reaction to C2 exceeds a threshold versus the percentage whose reaction to CI exceeds that same threshold, or comparing average levels of reactions to CI and C2 across the population), the researchers are interested in any patterns of results at the individual subject level. [00107] By plotting each subject's endpoint score for CI and C2 on an x-y graph, researchers can see that the change in chemistry reduces the impact on most of those subjects who had high levels of reactions to CI, but the change in chemistry exacerbates the adverse impact on some of them. Further, C2 produces an adverse reaction on a number of subjects who had no adverse reaction to CI .

[00108] This finding leads the researchers to analyze the genetic profiles of the subjects in the population to determine whether there is a genetic partem (such as different gene alleles) that distinguish those subjects who can tolerate CI better than C2, or vice versa. Any gene partem that is found can assist in further development of the chemistry, or the development of genetic tests, to determine the suitability of the specific formulation for subjects to whom the drug, or agent, will eventually be administered.

Example 4

[00109] A laboratory frequently uses samples of iPSCs derived from dermal fibroblasts from at least 10 subjects to assess a variety of endpoints when conducting in vitro tests for toxicity of agents, such as chemical compounds.

[00110] Previous tests of many agents have shown that the subjects in this population often respond significantly differently to identical doses of the same agent. That finding has been important in a number of decisions as to whether to proceed with the further development of the agent, as too large a distribution of responses may be an indicator that a certain subjects in the population are likely to have a severe adverse reaction to the agent in vivo, even if tests and clinical trials suggest that most subjects would not.

[00111] The protocol for conducting these tests involves creating (for each subject of the population) a bank of many vials of iPSCs that are aliquoted from a single colony of iPSCs grown from a single clone. When a test is ordered, one vial from each subject is thawed, and the cells from that vial are grown and expanded to produce the required number of cells to conduct the test. The success of that growth and expansion phase depends heavily upon the precise chemical composition and biological activeness of the reagents used during that period.

[00112] At a certain time, the researcher learns from the supplying manufacturer that an important reagent will no longer be available. Thus, the researcher will necessarily have to substitute for a reagent produced by another manufacturer, and that reagent may not have the identical chemical and biological properties as its predecessor. The researcher identifies two candidates - herein referred to as Reagent A and Reagent B.

[00113] In order to determine whether the results of future experiments can be directly compared to past experiments, the researcher must determine to a high degree of confidence that the substitution will not result in any significant differences in the biological behavior, function, and/or structure of cells treated with the candidate reagent versus those treated with the original reagent.

[00114] Precise comparative analysis of the chemical properties alone will not suffice, as reagents with seemingly identical properties can result in different biological behavior. This phenomenon is so common that scientists often insist that all iterations of experiments (e.g., across multiple subjects) be conducted using aliquots from a single manufacturing batch of a reagent, not even permitting the use of multiple batches of the same reagent from the same manufacturer.

[00115] Prior to development of the method described herein, researchers were limited in their ability to compare the impact of two (or more) reagents to the results of side-by-side experiments using one, or a very few cell lines. Given the underlying variability in the reactions of subjects to an identical challenge, this method could hide the potential for different reagents to affect portions of the population (if the one or a few cell lines did not include representatives of those particular portions).

[00116] Using the method described herein, the researcher begins by comparing reagent A to the original reagent. She conducts a series of paired experiments (involving a range of agents and endpoints) wherein all elements of the platform regimens and experiment regimens are held constant across both members of the pair, except for the use of the original reagent in one member of a pair and candidate reagent A in the other.

[00117] The two results (one from each member of the pair) for each subject of the population are used to create an x-y scatterplot diagram (with the result from the experiment using Reagent A determining a plot's position on the X axis experiment, and the result from the experiment using the original reagent determining a plot's position on the Y axis). She then conducts a number of statistical tests on the resulting set of data.

[00118] The results show that the co-efficient of correlation of the plots is less than 0.90, and that the slope of the linear regression line deviates from 1 : 1, particularly as the absolute values of the endpoint scores increases (within the relevant range of endpoint scores). The researcher rejects Reagent A as a viable substitute, given the importance of comparing the results of future experiments with the results of past experiments.

[00119] The researcher then repeats the procedure, using candidate Reagent B, which proves to be acceptable, as the co-efficient of correlation is 0.98, and the slope of the linear regression line remains at 1 : 1 throughout the relevant ranges of endpoint scores.

[00120] The researcher therefore chooses to use Reagent B in future experiments.

Example 5

[00121] A laboratory frequently uses samples of iPSCs derived from dermal fibroblasts from at least 10 subjects to assess a variety of endpoints when conducting in vitro tests for toxicity of agents.

[00122] The protocol for conducting these tests involves creating (for each subject of the population) a bank of many vials of iPSCs which were aliquoted from a single colony of iPSCs grown from a single clone. When a test is ordered, one vial from each donor is thawed, and the cells from that vial are grown and expanded to produce the required number of cells to conduct the test.

[00123] After some time, the existing supply of vials is nearing exhaustion. Given the cells involved are iPSCs, the possibility exists to take one vial (for each subject of the population) and expand the number of cells from that vial to an enormous quantity, in order to replenish the cell bank. However, the resulting cells would be the product of many more population doublings and passages than the cells in the vials created from the original colony.

[00124] Some research papers have suggested that the behavior of cells (including potentially, their reactions to challenge by stimuli such as agents, or chemical compounds) may change as the passage number grows.

[00125] Thus, to determine whether the "offspring" cells newly obtained after growth and expansion behave similarly to the original "parent" cells, the researcher determines to conduct side-by-side pairs of experiments analogous to those in Example 4. In this case, the pair is defined by a single change in the platform regimen (i.e., one member of the pair utilizes "parent" cells, while the other member utilizes "offspring" cells), while all elements of the experiment regimen are held constant across the pair. [00126] The two results (i.e. one from each member of the pair of experiments) for each subject of the population are used to create an x-y scatterplot diagram (with the result from the experiment using the "offspring" cells determining a plot's position on the X axis experiment, and the result from the experiment using the "parent" cells determining a plot's position on the Y axis). She then conducts a number of statistical tests on the resulting set of data.

[00127] The results prove to be acceptable, as the co-efficient of correlation is .98, and the slope of the linear regression line remains at 1 : 1 throughout the relevant ranges of endpoint scores.

[00128] The researcher therefore proceeds to refresh the bank by expanding one vial from each member of the cohort, and aliquoting the resulting cells into an appropriate number of vials.

Example 6

[00129] The discovery that fever (an easily and non-invasively measured biological reaction) is highly correlated with infection and inflammation (a biological reaction which is difficult to directly measure if that infection or inflammation is occurring in an internal organ) was valuable for patient diagnosis and treatment.

[00130] A researcher desires to investigate whether there might be certain easily measured biological responses to agents, or chemical compounds, that correlate with, or are predictive of, other key biological reactions that are difficult to directly measure in a living patient. For instance, the measurable response (e.g., response indicator) may constitute an in vitro test that is indicative, diagnostic, or predictive of an in vivo condition. Specifically, the researcher chooses to investigate whether there is a non-invasive predictor of "cardiomyocyte mitochondrial damage" phenomenon which occurs when an agent damages the mitochondria within the heart muscle cells (cardiomyocytes), thereby reducing that person's heart's ability to store and utilize energy, which can produce cardiac arrest. This phenomenon is difficult to directly measure in vivo, as direct measurement would require a biopsy of the patient's heart, a clearly invasive procedure.

[00131] However, in vitro measures of cardiomyocyte damage have been shown to bear correlation with in vivo mitochondrial damage. Therefore, if the researcher can establish a strong statistical linkage between mitochondrial damage in vitro and another endpoint in vitro - one that can also be directly and non-invasively measured in vivo - then he will have discovered a clinically relevant diagnostic tool.

[00132] The researcher has previously established a single platform regimen and experiment regimen, and conducted experiments that provided endpoint scores from a variety of assays across a number of agents (each at a variety of dose concentrations) for a population, while following the methods covered herein. Specifically, one of those assay endpoints was a score for the mitochondrial damage that results from the administration of an agent at a specified dose concentration level.

[00133] The researcher then selects one endpoint of a different assay that has been previously conducted on the cardiomyocytes. He examines only those in vitro endpoints whose scores have been successfully correlated with the endpoint scores of a non-invasive in vivo test. This endpoint is referred to herein as a Qualifying Endpoint.

[00134] The researcher conducts an analysis (herein referred to as Module 1) comprising the correlation of the x-y plots obtained for each subject of the population challenged by a single concentration of one agent by plotting: (1) the scores for mitochondrial damage, against (2) the score produced for a Qualifying Endpoint, and conducting the necessary statistical analysis.

[00135] The researcher then repeats Module 1 separately for all other dose concentrations of that same agent on which experiments have previously been conducted and recorded under this method described herein. This aggregation of multiple iterations of Module 1 is referred to herein as Procedure A.

[00136] Procedure A is then repeated across all other agents for which experiments have previously been conducted and recorded under the method described herein.

[00137] The researcher then conducts statistical analyses to determine whether there is a sufficiently robust statistical relationship between the two in vitro tests to conclude that the easier to perform test is a valid indicator of mitochondrial damage.

[00138] Procedure A is repeated for all other Qualifying Endpoints identified.

[00139] As a result, the researcher discovers one non-invasive in vivo test that can provide an indication of mitochondrial damage.

Example 7 [00140] A researcher uses the method described herein to discover aspects of the function of a gene.

[00141] A researcher conducts an assay of a number of agents, or compounds, on cardiomyocytes derived from at least 10 subjects under conditions dictated by the method described herein. Through this work, she discovers that a certain gene (referred to herein as Gene X), which is known to exist in two allele forms (Allele B and Allele B), produces a cytotoxic reaction, as measured by a particular endpoint, only when the donor has Allele A. The magnitude of the response (as measured by the endpoint) varies from subject to subject if Allele A is present, but is always zero when Allele B is present.

[00142] However, at the time of the research, science has not determined the function of Gene X, which could be cardiac-specific, or a function that pertains to some, many or all cell types in the body.

[00143] The researcher twice repeats the assay using the method described herein, holding all elements of the platform regimen and experiment regimen constant, with the exception that the researcher alters the functional cell type each time - once using hepatocytes (derived, of course, from the same iPSC cell banks from which the cardiomyocytes had previously been derived), and once using (similarly derived) neurons. While the tests are conducted on cells from all subjects, the subjects are analytically separated into two sub-populations, based on whether a given donor possesses Allele A or Allele B. Later analysis is conducted on these two sub-populations separately.

[00144] The researcher then uses the three data points (i.e., cytotoxicity score for cardiomyocytes, cytotoxicity score for hepatocytes, and cytotoxicity score for neurons) for each subject for each agent and dose concentration to conduct tests of correlation across the three variables (analyzing each sub-population separately). The results for the each agent- dose set of data consistently indicate that the endpoint scores for donors with Allele A are strongly correlated across all three cell types, while results for donors possessing Allele B are consistently zero (within a small error range).

[00145] This research suggests that the gene's function is likely to be one that influences the behavior of multiple cell types in the body, and that further research on the gene's function should focus on those aspects of gene function that are common across

cardiomyocytes, hepatocytes and neurons, rather than any that are idiosyncratic to any one cell type. [00146] Publications cited herein are hereby specifically incorporated by reference in their entireties and at least for the material for which they are cited.

[00147] Various embodiments of the invention have been described in fulfillment of the various objectives of the invention. It should be recognized that these embodiments are merely illustrative of the principles of the present invention. Numerous modifications and adaptations thereof will be readily apparent to those of skill in the art without departing from the spirit and scope of the invention.

[00148] Disclosed herein are methods which can be used with an array of materials useful for carrying out any one or more disclosed method. Where a method is disclosed and a number of modifications to the method are discussed, each and every combination and permutation of the method, and the modifications that are possible, are specifically contemplated unless specifically indicated to the contrary. Likewise, any subset or combination of these is also specifically contemplated and disclosed. This concept applies to all aspects of this disclosure. Thus, if there are a variety of additional steps that can be performed, it is understood that each additional step can be performed with any specific method step or combination of method steps of the disclosed methods, and in any order or permutation, unless otherwise indicated, and that each such combination or subset of combinations is specifically contemplated.

Claims

CLAIMS We claim:

1. A method for determining a cross-subject relationship between two or more response indicators in a population comprising: a. selecting a platform regimen comprising a set of variables that selects separate samples from a population of at least 10 subjects, wherein the samples comprise proliferative multipotent or pluripotent stem cells; b. selecting an experiment regimen comprising a set of controlled experimental conditions and at least one stimulus, wherein the at least one stimulus is applied to or combined with the samples to generate a response; c. assaying the samples for response to the at least one stimulus, wherein a response indicator comprises a quantified response endpoint to the at least one stimulus; d. determining for each sample an intra-subject relationship between two or more response indicators; and e. comparing in a cross-subject study the intra-subject relationships between the two or more response indicators to determine the cross-subject relationship between the two or more response indicators.

2. The method of claim 1, wherein a population of at least 20 subjects are selected.

3. The method of claim 1, further comprising differentiating the samples into one or more functional cell types.

4. The method of claim 1, wherein the subjects are human.

5. The method of claim 1, wherein the stimulus comprises an agent.

6. The method of claim 1, wherein the samples are assayed with a first stimulus comprising a first agent and a second stimulus comprising a second agent, and wherein a first response indicator comprises a quantified response endpoint to the first agent and a second response indicator comprises a quantified response endpoint to the second agent.

7. The method of claim 6, further comprising identifying a sub-population of subjects for which the second agent provides increased efficacy, reduced side-effects, or both compared to other subjects of the population.

8. The method of claim 7, further comprising comparing a genetic profile of at least one subject in the sub-population with a genetic profile of at least one subject of the population that is not within the sub-population.

9. The method of claim 8, further comprising identifying a genetic determinant correlated with the sub-population's response to the second agent.

10. The method of claim 6, wherein assaying the samples with the first agent provides structural information about the samples, and assaying the samples with the second agent provides functional or behavioral information about the samples.

11. The method of claim 1, wherein the stem cells of step a) are grown and expanded.

12. The method of claim 11, wherein growing and expanding the stem cells occurs prior to assaying the samples for response to the at least one stimulus.

13. The method of claim 12, wherein the stem cells are grown and expanded in a first medium comprising a first agent and separately in a second medium comprising a second agent.

14. The method of claim 13, wherein the first agent and the second agent are reagents of the growth and expansion medium.

15. The method of claim 13, wherein the comparing step e) determines the relationship between the first and second agents on growth and expansion of the stem cells.

16. The method of claim 12, wherein the samples assayed comprise a parental aliquot of each sample before growth and expansion, and an offspring aliquot of each sample obtained after growth and expansion.

17. The method of claim 1, wherein at least one of the two or more response indicators can be measured in an in vitro assay and minimally invasively in vivo.

18. The method of claim 17, wherein the method validates an in vitro test as being indicative, diagnostic, or predictive of an in vivo condition.

19. The method of claim 3, wherein the stem cells are differentiated into at least two different functional cell types, and wherein the at least two different functional cell types of each sample are assayed and compared.

20. The method of claim 19, wherein the intra-subject relationship comprises a relationship between two or more alleles of a gene.