WO2026039502A1

WO2026039502A1 - Assessing experimental reproducibility

Info

Publication number: WO2026039502A1
Application number: PCT/US2025/041768
Authority: WO
Inventors: Albert H. Gough; Mark E. Schurdak; Douglass Lansing Taylor
Original assignee: University of Pittsburgh
Current assignee: University of Pittsburgh
Priority date: 2024-08-15
Filing date: 2025-08-13
Publication date: 2026-02-19
Anticipated expiration: 2027-02-15

Abstract

Systems and methods for evaluating reproducibility and heterogeneity for in vitro models and studies based on in vitro models include a computer that may obtain experiment results data for a plurality of microphysiology system models of a plurality of microphysiology system studies. The results data may include a plurality of measurement data values at a plurality of interval points in a microphysiology system study. The computer may generate variability values based upon each respective measurement data value for the microphysiology system models at one or more interval points of the microphysiology system study. The computer may determine whether one or more variability values fail to satisfy corresponding reproducibility thresholds and generate a reproducibility status for the microphysiology system models or microphysiology system studies, indicating whether the microphysiology system models or microphysiology system studies are likely reproducible. The computer may then determine whether the variability values of certain reproducible microphysiology system models satisfy a heterogeneity threshold.

Description

076333-1039 / 06749 PATENT

ASSESSING EXPERIMENTAL REPRODUCIBILITY

CROSS-REFERENCE TO RELATED APPLICATION

[0001] This application claims priority to U.S. Provisional Application No. 63/683,663, filed August 15, 2024, which is incorporated by reference in its entirety.

STATEMENT OF GOVERNMENT FUNDING

[0002] This invention was made with government support under grant numbers: TR002632, DK117881, DK119973, TR003289, and TR001935, awarded by the National Institute of Health. The government has certain rights in the invention.

TECHNICAL FIELD

[0003] Method for assessing the reproducibility of experimental studies based on multicell physical models and distinguishing the reproducibility of the physical models from biologically/clinically relevant heterogeneity.

BACKGROUND

[0004] In vitro models of cells, tissues, organs, and organ systems have evolved as forms of physical models from simple, two-dimensional (2D), one-cell type experimental models to complex, three-dimensional (3D), multi-cell type models (including tissue slices), that support measurements over time and space. In particular, human microphysiology system (MPS) and organs-on-chip (OoC) modeling techniques have emerged as powerful platforms for recapitulating normal organ/organ system functions and disease progression, and are being actively applied in drug discovery and development programs with a goal of using them in precision medicine. Biomedical research often involves developing 2D cellular culture models or 3D multi-cellular type models, such as an MPS, to model or mimic normal organ functions and disease progression. Such MPS models are becoming increasingly popular and important tools in drug discovery and development to support precision medicine, among other benefits. An MPS includes advanced physical models that use different types of cells arranged in three dimensions to closely replicate the environment and functionality of human organs or organ systems. This technology allows researchers to observe how drugs interact with tissues and organs in a way that 2D cell cultures often cannot. One application of MPS models is in the selection of patient groups, or cohorts, for clinical trials. By using MPS models, scientists can identify which patients are most likely to

Page 1 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT benefit from a particular treatment, making clinical trials more efficient and targeted. Additionally, MPS are used in drug discovery and development and can help identify the best therapeutic strategies for specific patient groups, leading to more personalized and effective treatments.

[0005] A critical step in the use of MPS for these research benefits is demonstrating the reproducibility of these systems. “Reproducibility” refers to the ability to consistently replicate results across different experiments and studies when run identical conditions including the same cell samples. This is crucial for ensuring that the findings from MPS studies are reliable and can be trusted in the development of new drugs and treatments. Furthermore, it is essential to distinguish biological and clinical heterogeneity from experimental variability. “Heterogeneity” refers to the natural differences found within biological systems or patient populations, such as genetic variations or differences in disease states. “Experimental variability,” on the other hand, refers to variations that arise from the experimental procedures themselves, such as differences in how the experiments are conducted or how the data is collected. In order to properly interpret studies between patient MPS and experimental treatments, the experimental variability must be defined and fall within predefined ranges.

[0006] To achieve and demonstrate these qualities, researchers must capture detailed metadata associated with MPS studies, which includes the relevant information about the experimental reagents, cells, conditions, procedures, and data. Additionally, a strong analytical approach is necessary to assess reproducibility and ensure that the observed differences are due to biological or clinical heterogeneity rather than experimental variability. Evaluating and confirming the reproducibility of MPS models and distinguishing between biological heterogeneity and experimental variability are critical for successfully implementing MPS models in developing and evaluating drugs and providing precision medicine dedicated to particular patients.

SUMMARY

[0007] Disclosed herein are systems and methods capable of addressing technical shortcomings and may provide any number of additional or alternative benefits and advantages. In prior approaches to experiments involving 2D physical models many measurements and metrics may be statistical evaluations between points-in-time when the measurements are taken for the 2D model. The simple 2D models can use simple statistics, such as the Z’ metric, for accurate analyses. However, prior approaches are often too imprecise or inaccurate for evaluating reproducibility and

Page 2 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT heterogeneity for 3D MPS models used for MPS-based experiments and studies. Multi-cellular 3D MPS models have substantially more physical and spatial complexity, so determining and proving reproducibility and heterogeneity requires a computing system executing a multi-tiered, data driven, approach to designing and adjusting the MPS models and experiments using substantial amounts of data, measurement dependencies, and machine-learning architectures.

[0008] For ease of description and understanding, the embodiments described herein refer to MPS models and studies as physical models and experimental studies. However, embodiments are not limited to MPS models or MPS-based studies. The potential types of physical models and studies may include any cell-based in vitro model or related study, including nearly any form of 2D and 3D in vitro models and related studies (and not limited to MPS models and studies). It is understood that the term MPS used herein encompasses all such models including those involving static or microfluidic platforms of 3D models of cells as well as those comprised of self-assembly organoids and physically layered cells.

[0009] In an embodiment, a computer-implemented method for managing and analyzing in vitro data (associated with, for example, MPS study data) may include obtaining, by a computer from a database (e.g., an MPS database, such as EveAnalytics), experiment results data for a plurality of MPS models (which can include static or microfluidic platforms of 3D models of cells including self-assembly organoids and physically layered cells) of a plurality of MPS studies, indicating a plurality of measurement data values at a plurality of timepoints in an MPS study; generating, by the computer, a first single timepoint variability value based upon each respective measurement data value for the MPS models at a given timepoint of the MPS study; in response to determining that the first single timepoint variability value fails to satisfy a first reproducibility threshold, generating, by the computer, a second single timepoint variability value based upon each respective measurement data value for the MPS models at the given timepoint in each MPS study; and generating, by the computer, a reproducibility status for the MPS models based upon the first single timepoint variability value and the second single timepoint variability value.

[0010] In another embodiment, a system for analyzing cell-based in vitro data comprises a computer comprising at least one processor configured to: obtain, from a database (e.g., an MPS database, such as EveAnalytics), experiment results data for a plurality of MPS models (which can include static or microfluidic platforms of 3D models of cells including self-assembly organoids

Page 3 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT and physically layered cells) of a plurality of MPS studies, indicating a plurality of measurement data values at a plurality of timepoints in an MPS study; generate a first single timepoint variability value based upon each respective measurement data value for the MPS models at a given timepoint of the MPS study; in response to determining that the first single timepoint variability value fails to satisfy a first reproducibility threshold, generate a second single timepoint variability value based upon each respective measurement data value for the MPS models at the given timepoint in each MPS study; and generate a reproducibility status for the MPS models based upon the first single timepoint variability value and the second single timepoint variability value.

[0011] Embodiments may include a computer-implemented method for analyzing cellbased in-vitro data. The method may include: obtaining, by a computing device, experimenting results data associated with a plurality of microphysiology system constructs (sometimes referred to as “models”) of a plurality of microphysiology system experimenting operations sequences (sometimes referred to as “studies”), indicating a plurality of measurement data values at a plurality of interval points (or timepoints) in a microphysiology system experimenting operations sequence; generating, by the computer, a first single interval point variability value based upon each respective measurement data value for the microphysiology system at a given interval point of the microphysiology system experimenting operations sequence; in response to determining that the first single interval point variability value does not satisfy a first reproducibility threshold: generating, by the computer, a second single interval point variability value based upon each respective measurement data value for the microphysiology system at the given interval point in each microphysiology system experimenting operations sequence; and generating, by the computer, a reproducibility status for the microphysiology system based upon the first single interval point variability value and the second single interval point variability value.

[0012] The method may include generating, by the computer, a homogeneity variability value using the experiment results data for a first subset of microphysiology system constructs having a predetermined attribute value; and identifying, by the computer, a homogeneity in the first subset of microphysiology system constructs in response to determining that the homogeneity variability value satisfies a homogeneity threshold to assign a reproducibility status.

[0013] The method may include selecting, by the computer, the first subset of microphysiology system constructs including an identical set of cell attributes corresponding to a

Page 4 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT same subject (e.g., patient) and associated with a particular experimental treatment; selecting, by the computer, a second subset of microphysiology system constructs including a plurality of varied cell attributes associated with a plurality of subjects; and identifying, by the computer, a biological heterogeneity status for the first subset of microphysiology system constructs for the same subject distinct from an experimental variability for the first subset of microphysiology system constructs. The computer identifies the biological heterogeneity status in response to determining that the second subset of microphysiology system constructs includes the plurality of varied cell attributes for the plurality of subjects has a non-reproducibility status.

[0014] The method may include determining by the computer, whether the second single interval point variability value satisfies a second reproducibility threshold.

[0015] The method may include generating, by the computer, a first multi-interval point variability value for the microphysiology system constructs of the plurality of microphysiology system experimenting operations sequences based upon the plurality of respective measurement data values for the microphysiology system constructs at each interval point; and determining, by the computer, whether the first multi-interval point (or multi-timepoint) variability value for the microphysiology system constructs satisfies a first multi- interval point reproducibility threshold.

[0016] The method may include in response to determining that the first multi-interval point variability value fails to satisfy a the first multi-interval point reproducibility threshold: generating, by the computer, a second multi-interval point variability value for the microphysiology system constructs based upon the plurality of measurement data values for the plurality of microphysiology system constructs at the plurality of interval points in each microphysiology system experimenting operations sequence; and determining, by the computer, whether the second multi-interval point variability value for the microphysiology system constructs satisfies a second multi-interval point reproducibility threshold.

[0017] The method may include updating, by the computer, the reproducibility status for the microphysiology system constructs according to the first multi-interval point variability value and the second multi-interval point variability value.

[0018] The method may include executing, by the computer, a microphysiology system configuration engine taking a set of microphysiology system experimenting operations sequence

Page 5 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT parameters as input. The microphysiology system configuration engine having a machine-learning model trained to generate a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

[0019] The method may include re-training, by the computer, machine-learning model of the microphysiology system configuration engine in response to determining that the reproducibility status for at least one microphysiology system construct fails to satisfy a corresponding reproducibility status threshold.

[0020] The method may include generating, by the computer, a user interface presenting a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

[0021] When obtaining the experiment results data, the method may include receiving, by the computer from a laboratory testing device, the experiment results data associated with one or more microphysiology system constructs.

[0022] Embodiments may include a system for analyzing cell-based in-vitro data. The system may include a computer having at least one processor configured to: obtain, from a database, experiment results data for a plurality of microphysiology system constructs of a plurality of microphysiology system experimenting operations sequences, indicating a plurality of measurement data values at a plurality of interval points in a microphysiology system experimenting operations sequence; generate a first single interval point variability value based upon each respective measurement data value for the microphysiology system constructs at a given interval point of the microphysiology system experimenting operations sequence; in response to determining that the first single interval point variability value fails to satisfy a first reproducibility threshold, generate a second single interval point variability value based upon each respective measurement data value for the microphysiology system constructs at the given interval point in each microphysiology system experimenting operations sequence; and generate a reproducibility status for the microphysiology system constructs based upon the first single interval point variability value and the second single interval point variability value.

[0023] The computer may be further configured to: generate a homogeneity variability value using the experiment results data for a first subset of microphysiology system constructs

Page 6 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT having a predetermined attribute value; and identify a homogeneity status for microphysiology system constructs in response to determining that the homogeneity variability value for the subset of microphysiology system constructs satisfies a homogeneity threshold to assign a reproducibility status.

[0024] The computer may be further configured to: select the first subset of microphysiology system constructs including an identical set of cell attributes corresponding to a same subject (e.g., patient) and associated with a particular experimental treatment; select a second subset of microphysiology system constructs including a plurality of varied cell attributes associated with a plurality of subjects; and identify a biological heterogeneity status for the first subset of microphysiology system constructs for the same subject distinct from an experimental variability for the first subset of microphysiology system constructs. The computer identifies the biological heterogeneity status in response to determining that the second subset of microphysiology system constructs including the plurality of varied cell attributes for the plurality of subjects has a non-reproducibility status.

[0025] The computer may be further configured to determine whether the second single interval point variability value satisfies a second reproducibility threshold.

[0026] The computer may be further configured to: generate a first multi-interval point variability value for the microphysiology system constructs of the plurality of microphysiology system experimenting operations sequences based upon the plurality of respective measurement data values for the microphysiology system constructs at each interval point; and determine whether the first multi-interval point variability value for the microphysiology system constructs satisfies a first multi-interval point reproducibility threshold.

[0027] The computer may be further configured to: in response to determining that the first multi-interval point variability value fails to satisfy the first multi-interval point reproducibility threshold: generate a second multi-interval point variability value for the microphysiology system constructs based upon the plurality of measurement data values for the plurality of microphysiology system constructs at the plurality of interval points in each microphysiology system experimenting operations sequence; and determine whether the second multi-interval point

Page 7 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT variability value for the microphysiology system constructs satisfies a second multi-interval point reproducibility threshold.

[0028] The computer may be further configured to update the reproducibility status for the microphysiology system constructs according to the first multi-interval point variability value and the second multi-interval point variability value.

[0029] The computer may be further configured to execute a microphysiology system configuration engine taking a set of microphysiology system experimenting operations sequence parameters as input, the microphysiology system configuration engine having a machine-learning model trained to generate a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

[0030] The computer may be further configured to re-train the machine-learning model of the microphysiology system configuration engine in response to determining that the reproducibility status for at least one microphysiology system construct fails to satisfy a corresponding reproducibility status threshold.

[0031] The computer may be further configured to generate a user interface presenting a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

[0032] When obtaining the experiment results data, the computer may be further configured to receive, from a laboratory testing device, the experiment results data associated with one or more microphysiology system constructs.

[0033] It is to be understood that both the foregoing general description and the following detailed description are exemplary and explanatory and are intended to provide further explanation of the invention as claimed.

BRIEF DESCRIPTION OF THE DRAWINGS

[0034] The present disclosure can be better understood by referring to the following figures. The components in the figures are not necessarily to scale, emphasis instead being placed upon illustrating the principles of the disclosure. In the figures, reference numerals designate corresponding parts throughout the different views.

Page 8 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0035] FIG. 1 shows components of a system for therapeutics modeling, such as the MPS database (e.g., Eve Analytics), using a machine-learning architecture for processing various types of subject data, according to an embodiment.

[0036] FIG. 2 shows dataflow among devices of a system for MPS-model development and experimentation, according to an embodiment.

[0037] FIG. 3A shows the dataflow amongst the software components of the server performing a process for evaluating reproducibility and heterogeneity for a single timepoint metric, according to an embodiment.

[0038] FIG. 3B shows the dataflow amongst the software components of the server performing a process for evaluating reproducibility and heterogeneity for a multiple timepoint metric, according to an embodiment.

[0039] FIG. 4A shows dataflow amongst software components of a server of a system performing a process for evaluating reproducibility and heterogeneity of MPS models across a plurality of experiments of an MPS-based study, according to an embodiment.

[0040] FIG. 4B includes graphical user interfaces displaying data measurements and reproducibility status outputs generated for the evaluation process, according to an embodiment.

[0041] FIG. 5 shows the dataflow in a computer-executed process for determining reproducibility and heterogeneity when the MPS model is reproducible, according to an embodiment.

[0042] FIG. 6 is a flowchart of an example computer-implement method for managing MPS models and systems, according to an embodiment.

DETAILED DESCRIPTION

[0043] Reference will now be made to the illustrative embodiments illustrated in the drawings, and specific language will be used here to describe the same. It will nevertheless be understood that no limitation of the scope of the invention is thereby intended. Alterations and further modifications of the inventive features illustrated here, and additional applications of the

Page 9 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT principles of the inventions as illustrated here, which would occur to a person skilled in the relevant art and having possession of this disclosure, are to be considered within the scope of the invention.

[0044] Embodiments described herein provide for a hardware and software-based solution that evaluate reproducibility for MPS models (sometimes referred to as “microphysiology systems” or “microphysiology system constructs”) and/or evaluate heterogeneity. A computing device (e.g., server such as the MPS database now EveAnalytics) receives configuration parameters for MPS models and MPS-based studies (sometimes referred to as “experimenting operations sequences”), as well as experiment results data for the MPS studies. The configuration parameters include, for example, study-related metadata or experimental settings, among other types of information. The server may also receive instructions to perform functions for evaluating reproducibility and heterogeneity of MPS models and MPS studies using the experiment results data, the configuration parameters, attributes of patient cells (sometimes referred to as “subjects”), or other types of data for discriminating or segregating populations. The computer executes various functions and operations for computing certain measurement data values, measurement metrics, and variability values using the results data, where the metrics and variability values include, for example, a coefficient of variation (CV) value, an Analysis of Variance (ANOVA) value, and Intraclass Correlation Coefficient (ICC) value. The server may make decisions for evaluating the intra- and inter-study reproducibility of MPS models or study performance.

[0045] In embodiments, after the server determines that the MPS models or studies are shown to be reproducible when run under identical conditions, the server can be employed to identify biological/clinical heterogeneity. An integrative analytical database that includes searchable experimental data and metadata enables the server to segment the experimental results data based on any number of variables or attributes. The server (e.g., the MPS database platfomi or other novel platform) may, for example, parse datasets for populations of experiment results according to one or more variables or attributes until the server identifies a variability that satisfies a heterogeneity threshold value.

[0046] EXAMPLE SYSTEM COMPONENTS

[0047] FIG. 1 shows components of a system 100 for therapeutics modeling using a machine-learning architecture for processing various types of subject data, according to an

Page 10 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT example embodiment. The system 100 includes a therapeutics analytics system 101 having analytics servers 102 and analytics databases such as the MPS database 104 (sometimes referred to as a “federated database”), subject data databases 106a-106n (generally referred to as subject databases 106 or a subject database 106), conditioned subject data databases 108a-108n (generally referred to as conditioned databases 108 or a conditioned database 108), and user devices 114. One or more networks 112 interconnect the components of the system 100, allowing the devices to communicate with one another.

[0048] Embodiments may comprise additional or alternative components or omit certain components from those of FIG. 1 and still fall within the scope of this disclosure. It may be common, for example, to include multiple analytics servers 102. Embodiments may include or otherwise implement any number of devices capable of performing the various features and tasks described herein. For instance, FIG. 1 shows the analytics server 102 as a distinct computing device from the analytics database 104. In some embodiments, the analytics database 104 includes an integrated analytics server 102.

[0049] The system 100 includes one or more networks 112, which may include any number of internal networks, external networks, private networks (e.g., intranets, VPNs), and public networks (e.g., Internet). The networks 112 comprise various hardware and software components for hosting and conduct communications amongst the components of the system 100. Non-limiting examples of such internal or external networks may include a Local Area Network (LAN), Wireless Local Area Network (WLAN), Metropolitan Area Network (MAN), Wide Area Network (WAN), and the Internet. The communication over the networks 112 may be performed in accordance with various communication protocols, such as Transmission Control Protocol and Internet Protocol (TCP/IP), User Datagram Protocol (UDP), and IEEE communication protocols, among others.

[0050] The system 100 includes various hardware and software components of the analytics system 101. The analytics system 101 may include a computing network infrastructure comprising physically and logically related software and electronic devices, managed or operated by, for example, a therapeutic modeling service provider, where the devices of the infrastructure 101 are configured to provide the intended therapeutic modeling services. The analytics system 101 may include an internal network (not shown) comprising the networking hardware and

Page 11 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT software components for hosting and conducting the communications amongst components of the analytics system 101, including communications between the analytics server 102 and the analytics database 104.

[0051] The analytics system 101 may be hosted, implemented, operated, administered, or otherwise used by an organizational entity or researchers, such as a therapeutic organization (e g., hospital, physician) or research organization (e g., university, pharmaceutical company, research agency). The analytics system 101 includes one or more analytics servers 102 having hardware and software components that provide various features, functions, and benefits described herein, such as features and functions for therapeutics modeling, including developing and evaluating various types of therapeutic models. Non-limiting examples of therapeutic modeling include developing and evaluating physical MPS models, developing and evaluating clinical trials, developing and evaluating therapeutic strategies, and predicting and evaluating drug efficacy, among others.

[0052] For instance, in some cases, the analytics system 101 may develop or receive MPS configuration information containing various types of MPS information about the MPS models of the patients, where the configuration information indicates the types of data and/or the values corresponding to the types of configuration information. The analytics server 102 may receive configuration inputs from a user device 114 indicating the various types of MPS information for the MPS models and the related values for the particular MPS models. Alternatively, the analytics server may execute software programming to automatically develop, predict, or select MPS configuration parameters for automatically configuring the types of configuration information of one or more new MPS models based upon corresponding one or more experimental configurations entered and received from the user device 114. The analytics server 102 (or other component of the analytics system 101) may execute software functions to assess a patient’s attributes in the patient data from the subject databases 108 to predict, for example, subject-specific mechanisms of biological responses to certain study testing configurations, such as predicting disease progression or therapeutic strategies, predicting a cohort for testing a drug or treatment, or predicting a drug or treatment efficacy for a particular patient.

Page 12 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0053] In some cases, the analytics system 101 evaluates reproducibility of one or more MPS models for an MPS-based study. The analytics server 102 computes, generates, or otherwise determines a reproducibility of the MPS models.

[0054] Additionally or alternatively, in some cases, the analytics system 101 evaluates experimental variability in a manner that distinguishes the variability from heterogeneity. The analytics server 102 may compute, generate, or otherwise determine a homogeneity value or score of the MPS models across MPS models of the study or across multiple studies.

[0055] A clinician-user interacting with the analytics system 101 using the user device 114 may configure, develop, create, and reference the MPS models of the patients. The MPS models include physical testing biological material created by the clinician using samples that the clinician (or other actor) extracted from the subjects. The clinician or researcher prepares and conducts experiments on the MPS models to test or validate the predictions generated by or for a study (e.g., predicted therapeutic strategies).

[0056] The user device 114 allows the user (e.g., clinician, medical provider, researcher) to interact with the therapeutic modeling services of the analytics system analytics system 101. The user device 114 may include any computing device comprising hardware (e g., non-transitory machine-readable storage media, processors) and machine hardware-executed software components capable of performing the processes and tasks described herein. Non-limiting examples of the user device 114 may include a personal computer (PC) (workstation computer, laptop), tablet, and smartphone, among other types of electronic devices capable of performing the functions of the user device 114 described herein.

[0057] The user device 114 comprises, or couples to, peripheral devices for receiving the user inputs, such as user I/O devices (e.g., keyboard, mouse, monitor), allowing the user to interact with the user device 114 and the analytics system 101, via the network 112. As an example, the user may enter user inputs, such as configuration inputs indicating various types of configurations for developing or instantiating, for example, MPS-based studies, MPS models, or study evaluation functions (e.g., determining reproducibility, determining homogeneity), among others. In some cases, the user device 114 are coupled to peripheral devices, which may include instruments that obtain (e.g., receive, generate) experimental results data when conducting experimental analysis

Page 13 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT on the MPS models, and connectedly or wirelessly transmit the experimental results data to the user device 114 or to the analytics server 102. Non-limiting examples of such instruments include an image reader, microscope, live cell image reader, sequencing system, thermal cycler, and gel and blot readers. Additionally or alternatively, the user may enter the experimental results data into a user interface of the user device 114, which may in turn transmit the experimental results data to the analytics system 101 via the network 112.

[0058] The user device 114 executes various software programming for accessing the analytics system 101 via the one or more networks 112, allowing the user to provide user inputs to the analytics system 101 and receive outputs and data returned from the analytics system 101. In some implementations, the user device 114 executes locally installed software associated with the analytics system 101 for accessing and interacting with the services of the analytics system 101, and performing the various functions and features described herein. In some implementations, the user device 114 executes a web browser programing that accesses a website or web-app hosted by a webserver program executed by the analytics server 102 of the analytics system 101. The user may operate the web browser as the user interface for interacting with the services of the analytics system 101 and performing the various functions and features described herein.

[0059] The user may operate the user device 114 to submit requests for the analytics server 102 to perform the therapeutics modeling or evaluation functions, such as requests for predicted cohorts, requests for predicted therapeutics efficacy, requests for performing an evaluation of reproducibility, and/or requests for performing an evaluation of the homogeneity of the MPS model(s). The requests include machine-readable instructions that instruct the analytics server 102 to perform the requested activity, such as determining a reproducibility value, determining a homogeneity or heterogeneity value, predicting a cohort of subjects for testing a proposed therapeutic, or determining a therapeutics result for a proposed therapeutic, among other potential processes. The user device 114 receives the user inputs indicating the request(s) for the particular process(es) and/or various configuration inputs associated with the particular request(s) and then the user device 114 may transmit the request(s) to the analytics system 101 via the network 112.

[0060] The system 100 includes any number of subject databases 106 containing various types of subject data (sometimes referred to as “patient data”) for subjects (sometimes referred to as “patients”). A subject database 106 may be hosted on one or more computing devices

Page 14 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT comprising hardware components (e g., non-transitory machine-readable storage media, processor) and software components (e.g., database management system (DBMS)) capable of performing the various tasks and processes described herein. Similarly, a computing device may host one or more subject databases 106. In some cases, certain subject databases 106 may be a component of the analytics system 101. Additionally or alternatively, in some cases, certain subject databases 106 may be outside of the analytics system 101, and hosted by another enterprise infrastructure network, such that the analytics server 102 (or other component of the analytics system 101) ingests the subject data from the subject database 106 via the network 112. The subject databases 106 include various types of data used for generating or updating, for example, MPS- based study configurations, MPS configurations, experiment data, and/or various machinelearning models of one or more machine-learning architectures executed by the analytics server 102. Non-limiting examples of the subject databases 106 include clinical recordings database 106a, omics data 106b, MRI imaging 106c, and mobile health data 106d, among others. The types of data of the various subject databases 106 mentioned herein are merely examples and not intended as being limiting on potential embodiments.

[0061] A clinical database 106a may include demographics data indicating the subject’s age, gender, race, and ethnicity. The clinical database 106a may include medical history and comorbidities data indicating a subject’s disease history and surgical history. The clinical database 106a may include body composition data indicating a subject’s height, weight, fat mass, and the like. The clinical database 106a may include medications data indicating a subjects current and past medications. The clinical database 106a may include biochemistries data to indicate a subject’s blood chemistry. The clinical database 106a may include patient reported outcome measures to indicate a subject’s self-reported depression, anxiety, social function, or the like. The clinical database 106a may include noninvasive liver fibrosis staging to indicate a subject’s liver stiffness measurements and controlled attenuation parameters. The clinical database 106a may include liver biopsy data to indicate a subject’s steatosis grade score, inflammation score, ballooning score, and fibrosis stage. The clinical database 106a may include salivary and stool data to indicate a subject’s taxonomic classification and species abundance. The clinical database 106a may include plasma and serum data to indicate a subject’s metabolite and proteomic profiles. The clinical database 106a may include genetics data to indicate a subject’s exome sequencing profile.

Page 15 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0062] An omics database 106b may include genome data indicating DNA alterations that cause disease, transcriptome data indicating a subject’s RNA regulation profile, metabolome and lipidome data indicating a subject’s metabolite levels, metabolic and inflammatory proteome data indicating a subject’s serum inflammatory and metabolic protein levels, and spatial metabolomics indicating a subject’s metabolite counts per spot data. Non-limiting examples of types of data stored in the omics database 106b (or other subject databases 106) include clinomics data, radiomics data, genome data, and transcriptome data, among others.

[0063] The imaging database 106c may include data files and metadata generated by imaging devices, such as MRI image files generated by an MRI device (not shown). The image database 106c may include other types of data related to the imaging data, such as radiomics data to indicate, for example, a subject’s hepatic and body composition, cross-sectional abdominal imaging data to indicate a subject’s liver fat fraction, liver stiffness score, presence of portal hypertension, or presence of cirrhosis.

[0064] The mobile health database 106d may include patient reported medical data on portable devices, such as mobile phones, tablets, or laptop computers. Non-limiting examples include a subject’s vital signs, weight loss, sleep patterns, and physical activity.

[0065] In some embodiments, as in the example system 100, the analytics system 101 includes an analytics database 104 for storing information related to MPS-based studies and corresponding configurations. For example, the analytics database 104 includes configuration parameters for a given MPS-based study or configurations parameters for the types of data expected for the MPS models as developed and configured for a given study. As another example, the analytics database 104 includes configuration parameters for reproducibility evaluation functions and/or heterogeneity evaluation functions. For instance, the analytics server 102 (or other component of the system 100) executes the software functions for determining reproducibility or heterogeneity according to the configuration parameters entered at the user device 114 and stored in the analytics database 104, which may include types of data or metrics, data records of current or prior MPS-based studies, and one or more threshold values, among other configuration parameters.

Page 16 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0066] In some embodiments, as the analytics database 104 functions as a federated database storing the multimodal subject data from the plurality of subject databases 106. The analytics server 102 may receive and store the subject data obtained from the various subject databases 106 into the analytics database 104. For ease of explanation, the system 100 includes the analytics database 104 from which the analytics server 102 obtains subject data to perform the various processes described herein, though potential embodiments need not include such a central or federated database.

[0067] Optionally, the system 100 includes one or more conditioned databases 108. The conditioned databases 108 contain conditioned data, which includes preprocessed or “cleaned” instances of the subject data from the subject databases 106. The analytics server 102 or other computing device may execute any number of data conditioning processes, including preprocessing functions that clean, normalize, and format the subject data from the subject databases 106 for downstream operations. Non-limiting examples include data completion functions, data noise reduction functions, data transformation functions, and data normalization functions, among others. The output of such data conditioning processes applied to the subject data may be stored into the one or more conditioned databases 108 and/or into the analytics database 104. The example system 100 includes conditioned databases 108 corresponding to the subject databases 106, though embodiments need not include such a correspondence.

[0068] The analytics database 104 of the example system 100 may function as the federated database storing the multi-model subject data from the subject databases 106 and/or the conditioned databases 108. The analytics server 102 may receive and store the subject data obtained from the various conditioned databases 108 into the analytics database 104. For ease of explanation, the system 100 includes the analytics database 104 from which the analytics server 102 obtains the subject data from the subject database 106 and conditioned database 108 to perform the various processes described herein.

[0069] The analytics server 102 of the analytics system 101 may be any computing device comprising one or more processors and software, and capable of performing the various processes and tasks described herein. The analytics server 102 may host or be in communication with the analytics database 104, and receives and processes subject data, experimental results data, experimental configurations, and other types of inputs, which the analytics server 102 may receive

Page 17 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT from the subject databases 106, conditioned databases 108, and user devices 114, among other potential components of the system 100. Although FIG. 1 shows only a single analytics server 102, the analytics server 102 may include any number of computing devices. In some cases, the computing devices of the analytics server 102 may perform all or portions of the processes and benefits of the analytics server 102. The analytics server 102 may comprise computing devices operating in a distributed or cloud computing configuration and/or in a virtual machine configuration. In operation, the analytics server 102 may obtain (e.g., receive, retrieve) the subject data and/or the experiment results data from the various data sources (e.g., user devices 114, subject databases 106, conditioned databases 108, external databases, external websites) according to various preconfigured operations or instructions, and store such obtained data into the analytics database 104 for reference by the analytics server 102 when executing various downstream functional processes of the analytics server 102.

[0070] The analytics server 102 may automatically retrieve or receive the subject data from the subject databases 106 and/or the conditioned databases 108 via the networks 112, at a preconfigured interval or when one or more subject databases 106 or conditioned databases 108 are updated. Additionally or alternatively, the analytics server 102 may automatically retrieve or receive experiment results data from the analytics database 104 or other data sources (e.g., user devices 114, subject databases 106, conditioned databases 108, external databases, external websites) via the networks 112, at a preconfigured interval or when one of the other types of data sources (e.g., subject database 106, conditioned database 108) are updated. The clinician-user, researcher-user, the subject-user, or another type of actor may operate and instruct the user device 114 to manually upload or submit, via the networks 112, the subject data or experiment results from the subject database 106 or conditioned database 108 to analytics system 101, where the analytics server 102 or other computing device hosting the analytics database 104 may store the received subject data into the analytics database 104. The analytics server 102 may, for example, host an online precision-medicine portal allowing users to upload or otherwise submit the subject data and/or experiment results to the analytics system 101 for storage into the analytics database 104 or other database of the system 100.

[0071] In some implementations, the analytics server 102 may reference the various types of subject data and/or experiment results data, as stored and contained in analytics database 104

Page 18 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

(or other database(s) of the system 100) in order to train one or more machine-learning models of one or more machine-learning architectures. In such implementations, the analytics server 102 may execute software programming for a computational software module (sometimes referred to as a “computation module” or “computational module”) that generates, develops, and/or executes predictive machine-learning models. As another example, the predictive models generated, trained, or otherwise developed by the computational module may include cohort prediction engines for predicting selected cohorts for clinical trials. As another example, the computational module (or other software programming) of the analytics server 102 may generate or develop the configuration parameters for MPS models using MPS model-related data within the subject data or experiment results data (of one or more MPS-based studies) in the analytics database 104.

[0072] In some cases, the computational module or other software programming of the analytics server 102 may include and execute various layers of a machine-learning architecture among other functions. The computation module may implement various machine-learning and artificial intelligence algorithms that ingest and integrate the preprocessed data into various predictive models (e.g., cohort prediction engine, MPS model configuration prediction engine). The computational module may also ingest the MPS model-related data, such as experiment results data, for developing or updating the various types of predictive models. In some cases, the computational module or other software programming of the analytics server 102 executes programming for performing handcrafted models for generating or further developing a predictive model. In some implementations, the computational model employs and fuses both machinelearning models and handcrafted models to generate and develop certain predictive models. Nonlimiting examples of the models or techniques may include mechanistic models, stochastic models, and Bayesian networks, among other possible types of machine-learning models or handcrafted models.

[0073] As an example, the analytics server 102 may execute software programming of layers of a machine-learning architecture providing functions of a biomarker discovery module (sometimes referred to as a “biomarker module”). The analytics server 102 may train the biomarker module to identify potential clinical biomarkers from the subject data of any number of subjects and/or MPS model-related data, such as the experiment results data. In this way, the biomarker module may identify biology attributes of the subject in the analytics database 104 for creating the

Page 19 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

MPS models. Additionally or alternatively, the biomarker module (or other programming for a machine-learning model of the analytics server 102) may identify discriminating features for training and applying the various types of machine-learning models described herein.

[0074] The analytics server 102 may execute software programming layers of a machinelearning architecture including a machine-learning model functioning as a cohort prediction engine or cohort engine. The layers of the cohort engine may be trained to identify or predict a cohort of one or more subjects based on, for example, the subject data from the subject database(s) 106, the therapeutic testing configurations, and/or the MPS model-related data. In this way, the analytics system 102 may improve likely success for precision medicine outcomes, drug discovery and development outcomes, and clinical therapeutic treatment outcomes for individual subjects. For example, the analytics server 102 may predict the cohort of subjects having subject data indicating subjects having, for example, common major genetic attributes, lifestyle attributes, and environmental attributes; where the predictions are made on certain testing configuration, such as safety requirements, drug efficacy requirements, and drug candidate efficacy requirements.

[0075] In some implementations, the analytics server 102 may iteratively train layers of a machine-learning architecture for machine-learning architecture of an experiment selector for MPS-model based studies. The analytics server 102 may iteratively train the experiment selector to iteratively select experimental analysis constraints (e.g., experiments for MPS model-based studies) to efficiently construct a comprehensive map of drug efficacy within a cohort of users. The analytics server 102 may be trained to optimally prioritize the MPS-based experiments that would make the greatest contribution to the understanding of drug safety and efficacy. The analytics server 102 and/or the clinician may automatically or manually select or indicate experiment or MPS configuration design information (as testing configurations reflecting the experiment design or the MPS configuration parameters) for conducting the experimental analysis (or experiment) and for creating the MPS models.

[0076] As mentioned, following the MPS-based experiment or at predetermined intervals, the analytics server 102 obtains the experimental results data for the MPS models from one or more data sources (e.g., peripheral devices or measurement instruments, analytics database 104, user device 114), and then stores the experiment results data as the MPS model -related data for the corresponding MPS models into the analytics database 104 or other database(s) 106, 108. The

Page 20 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT programming of the analytics server 102 (e.g., computational module) may re-train or calibrate one or more predictive models of the analytics system 101 using the MPS model data, which includes the recent MPS-based experiment results data. This active learning approach may beneficially generate improved predictive machine-learning models or improve the predicted and suggested configurations of the MPS models. For instance, a loss layer of the machine-learning architecture (or other type(s) of computational model(s) executed by the analytics server 102) includes a loss function. The loss function determines a level of error of the computational model by determining a distance between the predicted output (e g., predicted experimental outcome, predicted cohort), against the corresponding observed outcome generated by the predictive machine-learning model (e.g., cohort prediction engine, MPS configuration prediction engine), as indicated by the experimental results data or other types of data. The loss function may tune the hyper-parameters or weights of the particular machine-learning model based upon the level of error, where the analytics server 102 may retrain the particular machine-learning model by executing the machine-learning model using the subject data and/or experiment results data and the level error, among other potential inputs.

[0077] EXAMPLE MODEL DEVELOPMENT

[0078] FIG. 2 shows dataflow among devices of a system 200 for MPS-model development and experimentation using predictive machine-learning models, according to an embodiment. The system 200 includes a therapeutics modeling and analytics system 201 (e.g., analytics system 101) having analytics servers 202 (e.g., analytics servers 102) and analytics databases 204 (e.g., analytics databases 104) for developing and evaluating MPS-based studies, MPS models (generated on microfluidic chips or other static 3D platform 210 such as plates), and/or predictive machine-learning models. The system 200 further includes hardware and software components for creating the physical MPS models according to configurations parameters and other inputs from the analytics server 202, including perfusion modules 208, a cell incubator 212, and an inline imaging reader 214 or other laboratory testing device, among other components. Embodiments may include or otherwise implement any number of devices capable of performing the various features and tasks described herein. For instance, FIG. 2 shows the analytics server 202 as a distinct computing device from the analytics database 204. In some embodiments, the analytics database 204 includes an integrated analytics server 202.

Page 21 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0079] The system 200 includes perfusion modules 208 housing any number of microfluidic chips or other static 3D platforms 210 (sometimes referred to as “organ chips,” “MPS chips,” or the like). As an example, a microfluidic chip 210 includes any number of chambers 206a-206c (generally referred to as chambers 206 or a chamber 206). The system 200 further includes the cell incubator 212 that receives the perfusion modules 208 and the inline imaging reader 214 coupled to the cell incubator 212 via a perfusion controller. Moreover, FIG. 2 depicts the dataflow for configuring an experimental analysis (or MPS-based experiment) using, for example, predictive models (e.g., cohort predictor, MPS-based experiment configuration predictor or selector), configuration parameters, and other types of data; the dataflow for performing the experiment using the MPS models of the microfluidic chips or other static 3D platforms 210 created according to the configuration parameters of the MPS experiment and the MPS models of the microfluidic chips or other static 3D platform 210; and the dataflow for capturing the experiment results data as generated from running the MPS-based experiment. The analytics system 201 is employed on the frontend of the dataflow to design the experiment, such as developing and executing predictive models for developing, configuring, and/or selecting the MPS models for the of the microfluidic chips or other static 3D platform 210 and other relevant information, such as subject data from the analytics database 204 or other types of data from other databases.

[0080] In component (A), the analytics server 202 may retrieve various types of data from the analytics database 204, such as predictive models, subject data, and metadata, among others; and may proceed to execute the predictive models. In some implementations, the analytics server 202 executes a cohort predictive model for predicting and selecting the cohort of subjects for the experiment. In some implementations, the analytics server 202 executes experiment selector model to identify a type of experiment to perform given the clinician’s testing configurations and the subject data for the subjects. Optionally, the machine-learning model layers of the experiment configuration engine may be trained to identify and output predicted or suggested configuration parameters of the MPS models for the microfluidic chips or other static 3D platforms 210 (e.g., plates) or MPS-based experiment.

[0081] In components (B)-(C), the clinician-user uses the outputs of the analytics system 201, including information related to the MPS design for the microfluidic chips or other 3D

Page 22 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT platforms 210, for setting up and performing the MPS-based experiment. The clinician may take samples from the cohort of subjects and prepare the microfluidic chips or other 3D platforms 210 according to the information received from various data sources (e.g., analytics database 204) of the analytics system 201. For instance, the clinician may take samples, such as cells, from the subjects for creating the MPS models or organoids of the subjects on one or more microfluidic chips or static 3D platforms such as plates 210.

[0082] In the example embodiment, a microfluidic chip or other static 3D platform 210 includes one or more culture chambers 206, and each chamber 206 comprises a perfusion channel and an injection port. When setting up the MPS model for the microfluidic chip or other static 3D platform 210, the subject’s cells are introduced through the injection ports or added prior to assembly of the microfluidic chip 210. Bubble traps are positioned on fluid paths to prevent air from entering the culture chambers 206. The chambers 206 provide luminal perfusion, basolateral perfusion, and cell tubules.

[0083] In component (D), the clinician introduces the microfluidic chip or other static 3D platform 210 to a perfusion module 208, which houses one or more microfluidic chips or other static 3D platforms 210 containing one or more corresponding MPS models. In addition, the perfusion module 208 includes fluid reservoirs and contains a bio layer, a reservoir layer, and a Pneumatic or hydraulic pressure layer.

[0084] In component (E), the perfusion module 208 containing the microfluidic chips or other static 3D platforms 210 are stored in a cell incubator 212, such as a CO2 cell incubator 212, containing one or more perfusion modules 208. The cell incubator 212 contains storage racks (e.g., 8 x 4 storage racks) for the perfusion modules 208, as well as a pneumatic or hydraulic pressure gas pump located outside the cell incubator 212. A perfusion controller, coupled to the cell incubator 212, manages the functions of the cell incubator 212.

[0085] In an example, a single microplate-footprint module 208 may be divided into 4 sections of 3 experiments (e.g., 4 microfluidic chips 210, each with 3 chambers 206 for cells), thereby supporting high-throughput of 12 experiments per well plate. Further, the integrated perfusion controller allows fluid passage to be controlled in each of the 4 microfluidic chips or other static 3D platforms 210. The pneumatic or hydraulic pressure control system of the cell

Page 23 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT incubator 212 may include high-resolution digital pressure controllers, along with on-chip flow restrictors and passive mixing channel, which produces user-programmable drug concentrations, for up to, e.g., 768 MPS models on microfluidic chips or other static 3D platforms 210 within the single cell incubator 212. The microfluidic chip or other static 3D platform 210 is configured to minimize drug absorption losses. The microfluidic chip or other static 3D platform 210 may be constructed from injected molded hard plastics and, in some embodiments, is not constructed with any silicone. In addition, in some embodiments, the microfluidic chip or other static 3D platform 210 and/or the cell incubator 212 includes interfaces for relatively easily coupling to confocal high- content imaging systems 214.

[0086] In component (F), a laboratory instrument, which may be coupled to the analytics server 202 or other computing device of the system 200, captures or generates the experiment results data for the MPS-based experiment, based upon the outputs of the cell incubator 212. For example, an imaging reader 214 (e.g., inline imagining reader) associated with the cell incubator 212 generates the experiment results data containing, for example, live cell readouts. Additionally or alternatively, the clinician may input the experimental results data to the user device (not shown).

[0087] In some cases, the MPS-based testing platform components may beneficially assess each MPS model rapidly in multi-dose, multi-drug treatment to evaluate a proposed therapeutic efficacy and safety. The MPS related data (e.g., MPS configuration preparation data, experiment results data) will include or indicate, for example, secretome, live cell, metabolomics or RNA-Seq, and/or endpoint IF imagery. This MPS -related data may be uploaded to the analytics system 201. In some cases, the MPS-related data may be combined with clinical measurements and omics data for upload to the analytics system 201.

[0088] In component (G), the user device or laboratory instrument (e.g., inline imaging reader 214) may transmit, upload, or otherwise provide the experiment results data to the analytics system 201. The analytics server 202 may receive and store the results data into the analytics database 204. The analytics server 202 may perform various downstream operations using the experiment results data, subject data, or other types of data.

Page 24 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0089] In some implementations, the analytics server 202 uses the experiment results data to train, re-train, tune, or otherwise update the predictive machine-learning models (e.g., cohort prediction engine, MPS-model configuration engine, MPS-model selection engine, MPS-based experiment configuration engine), which the analytics server 202 may execute to design or prepare the MPS-based experiments or MPS models. The predictive machine-learning models may include loss functions for adjusting the machine-learning models. When the analytics server 202 executes the loss functions, the analytics server 202 determines a level of error between a predicted output generated by the particular predictive machine-learning model and certain observed values in the experiment results data. In some cases, after the analytics server 202 receives the experiment results data, the analytics server 202 may further apply the one or more loss functions of the relevant predictive models to determine the corresponding level of error. For certain predictive models, the loss function may include programming for re-training or tuning the predictive model by adjusting or tuning the hyper-parameters or weights of the predictive model based upon, for example, the level of error, experimental results data, and subject data, among other potential types of data. In training, the analytics server 202 may determine that that any of the predictive models described herein are satisfactorily trained or output a sufficient result if the level of error satisfies a training threshold, a similarity threshold, or other threshold value.

[0090] The MPS-related data (e g., MPS configuration preparation data, experiment results data) includes, for example, secretome, live cell, metabolomics or RNA-Seq, and/or endpoint IF imagery data. This MPS-related data may be uploaded to the analytics system 201. In some cases, the MPS-related data may be combined with the clinical measurements and the omics data for upload to the analytics system 201. The analytics server 202 may incorporate the MPS- related data into the predictive machine-learning models that the analytics server 202 executed when designing the MPS-based experiment performed in the system 200.

[0091] EXAMPLE DATAFLOWS AND PROCESSES

[0092] FIGS. 3A-3B show dataflows amongst software components of a server 302 (e g., analytics server 102) of a system performing processes 300a, 300b for evaluating reproducibility and heterogeneity of MPS models and an MPS-based study or experiment. As mentioned, embodiments are not limited to MPS models and MPS-based studies. The server 302 executes the processes 300a, 300b to compute, generate, or otherwise determine the

Page 25 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT reproducibility MPS models using a preconfigured metric at one or more timepoints (or “interval point” at given time intervals) in experiment results data or other types of data. The server 302 may further execute the processes 300a, 300b to compute, generate, or otherwise determine heterogeneity using the preconfigured metric at the one or more timepoints in the experiment results data or other types of data, such that the server 302 may distinguish experimental variability from biological or clinical heterogeneity in the MPS models.

[0093] For ease of description, the server 302 executes the functions and features of the processes 300a, 300b, though embodiments are not so limited. Various types of computing devices may perform these processes 300a, 300b. Moreover, multiple computing devices may perform the features and functions mentioned in the descriptions of FIGS. 3A-3B.

[0094] The server 302 obtains (e.g., receives, retrieves) subject data and experiment results data from one or more data sources (not shown) of the system. The data sources may include client computing devices of end-users (e.g., user device 114) or one or more databases of the system (e g., analytics database 104, subject databases 106, conditioned databases 108), among other types of sources of subject data or experiment results data accessible to the server 302. The server 302 executes the various functions for evaluating reproducibility and heterogeneity in response to an instruction to invoke or otherwise execute the particular software functions. The server 302 may receive a request containing the instruction from a client device or input from a graphical user interface. In this way, an end-user may manually request and instruct the server 302 to evaluate the reproducibility and heterogeneity of the MPS models. Additionally or alternatively, the server 302 may receive configuration inputs that preconfigure the server 302 with the instruction to invoke the functions at a predetermined interval or in response to receiving updated data (e.g., updated subject data, updated experiment results data, updated experiment configuration parameters, updated MPS model configuration parameters). In the way, the software of the server 302 may be preconfigured with the instruction to automatically invoke the functions for evaluating reproducibility and heterogeneity for the MPS models.

[0095] In the example embodiments described in FIGS. 3A-3B, the server 302 implementing the processes 300a, 300b performs various functions for evaluating the reproducibility and heterogeneity, both intra-study and inter-study, using predetermined metrics and predetermined functions. To evaluate the intra- and inter-study reproducibility of experimental

Page 26 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

MPS-model performance, the functions performed by the server 302 may include computing, for example, Coefficient of Variation (CV), Analysis of Variance (ANOVA), and Intraclass Correlation Coefficient (ICC), among others.

[0096] Generally, during an MPS-based experiment, the server 302 or other device of the system 300 can capture measurement or observation data at any number of discrete, single points- in-time (e.g., value on Day 1 ; value on Day 3; value on Day 6). From the many discrete timepoints, the server 302 may take a particular instance of the measurement data values captured for a particular single timepoint and use this measurement data value as the single timepoint metric captured for that MPS model and/or that experimental condition of the MPS-based experiment. Additionally or alternatively, the server 302 can use multiple measurement data values of multiple timepoints (sometimes referred to as “multi-timepoint” or “multi-interval point”) to compute a collective metric using the multiple measurement data values from the time series.

[0097] As an example, a researcher may compute or identify an amount of collagen 1A1 levels secreted from Liver Acinus MPS (LAMPS) models of one or more studies, as the measurements or metrics taken at several different days over the course of the study, where each day is a timepoint. The server 302 receives inputs from the user interface or measurement instrument indicating, for example, an amount of collagen 1A1 secreted from a LAMP model on Day 2 (as timepoint 1), Day 4 (as timepoint 2), Day 6 (as timepoint 3), and Day 8 (as timepoint 4), where each value represents an absolute value of the metric for the signal timepoint. The server 302 may generate or receive the single timepoint metrics by executing one or more functions that generate values (e.g., reproducibility values, similarity values, variability values) that effectively compared what happened (e.g., measurement values) at timepoint 1 (e.g., Day 2), timepoint 2 (e.g., Day 4), and timepoint 3 (e.g., Day 6).

[0098] FIG. 3A shows the dataflow amongst the software components of the server 302 performing a computer-implemented method or device-executed process 300a for evaluating reproducibility and heterogeneity for a single timepoint metric. The server 302 executes the process 300a to compute, generate, or otherwise determine the reproducibility and/or heterogeneity for the MPS models using the preconfigured metric at a single timepoint in experiment results data.

Page 27 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0099] In operation 301, the server 302 obtains the single timepoint metric from the subject data or the experimental results data for the MPS models.

[0100] The single timepoint metric includes a type of measurement or other type of data as required by an evaluation request, instruction, or a configuration parameter for the particular MPS model or study. For example, on Day 8 of the MPS-based experiment, the server 302 may obtain or capture a particular measurement for the MPS model and update the experiment results data as captured for the MPS experiment. In this example, the single timepoint metric includes the particular measurement data value of the type of measurement captured for the MPS model on Day 8.

[0101] In operation 303, for the single timepoint metric, the server 302 computes a variability value or reproducibility value, and determines whether that variability value or reproducibility value satisfies one or more preconfigured reproducibility thresholds.

[0102] For instance, in the example process 300a, the server 302 first computes the CV (or other type of variability value) of the means of the replicate MPS models, samples, and/or MPS- based studies. The CV indicates the variability in the single timepoint metric’s values (e.g., variability of signal magnitude) amongst the replicate MPS models or samples within the MPS-based study (or “intra-study”). Additionally or alternatively, the CV may indicate the variability in the metric values (or “signal magnitude”) amongst the replicate MPS studies (or “inter-study”).

[0103] The server 302 then determines whether the CV satisfies one or more preconfigured reproducibility thresholds (e.g., CV < 5%; CV < 15%; CV > 15%), corresponding to a level of reproducibility or reproducibility status (e.g., excellent, acceptable, poor). For instance, the CV indicates the variability in the metric signal’s magnitude amongst the replicate samples (intra- study) and amongst the replicate studies (inter-study). In this example embodiment, a CV < 15% is preconfigured as indicating an “acceptable” or “excellent” (CV < 5%) reproducibility status across MPS-related samples or studies. Moreover, a CV > 15% indicates “poor” reproducibility status among the replicate samples, for both inter-study and intra-study analysis.

[0104] If the server 302 determines that the CV satisfied a minimally acceptable threshold reproducibility value or an “excellent” (or highest) threshold reproducibility value and status (in

Page 28 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT prior operation 303), then the server 302 may proceed directly to later operation 307. Otherwise, the server 302 proceeds to operation 305.

[0105] In operation 305, if the server 302 determined that the CV failed to satisfy the minimally acceptable threshold reproducibility value (in prior operation 303), then the server 302 performs one or more operations for computing secondary or other forms of variability values across multiple MPS studies. The server 302 determines whether the additional types of variability values satisfy another reproducibility threshold score.

[0106] In some implementations, the server 302 may execute functions that output a similarity or probability score (e.g., p-value) as the other type of variability value, where the probability score indicates a likelihood that the MPS studies are similar and reproducible. Generally, a higher p-value may indicate that the data points are similar and the more likely reproducible for the MPS model and/or MPS studies for the MPS-based experiments. The server 302 may determine whether this similarity or probability score satisfies probability or similarity threshold for the reproducibility.

[0107] For instance, when the CV >15% for inter-study comparisons, the server 302 performs an ANOVA function. The ANOVA compares the variability across studies against the variability within the studies as an indicator of differences among the groups of MPS models. The output of the ANOVA function includes the probability (p-value) that the studies are similar and reproducible. As an example, a p-value < 0.05 may signify a relatively low (5% or less) probability that the MPS studies are similar and thus are likely not reproducible. As another example, a p-value > 0.05 may signify a sufficient (greater than 5%) probability that the MPS studies are similar and thus are sufficiently likely reproducible.

[0108] In operation 307, the server 302 determines and generates a reproducibility status output. When the server 302 determines that the CV satisfied a minimally acceptable threshold reproducibility value or an excellent (or highest) threshold reproducibility (in prior operation 303), then the server 302 determines the reproducibility status is “acceptable” or “excellent.” The server 302 may further perform certain downstream operations based upon determining whether the similarity or probability score satisfied or failed the other reproducibility threshold (in prior

Page 29 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT operation 305). The server 302 stores the various types of experiment results data and reproducibility status output into one or more databases.

[0109] The server 302 executes or performs one or more downstream operations in accordance with the reproducibility status output. The server 302 may generate, for example, one or more messages for display on a graphical user interface for the clinician-user, indicating the reproducibility status. The server 302 may use the experiment results data and other reproducibility status outputs for training or retraining one or more machine-learning models.

[0110] FIG. 3B shows the dataflow amongst the software components of the server 302 performing a process 300b for evaluating reproducibility and heterogeneity for a multiple timepoint metric. The server 302 executes the process 300b to compute, generate, or otherwise determine the reproducibility and/or heterogeneity for the MPS models using one or more preconfigured metrics at a plurality of timepoints in experiment results data.

[0111] In operation 311, the server 302 obtains the multiple timepoint metrics from the subject data or the experimental results data as computed for the MPS models using the measurement value data across multiple moments-in-time.

[0112] In operation 313, for the multiple timepoint metrics, the server 302 computes a variability value or reproducibility value and determines whether that variability value or reproducibility value satisfies one or more preconfigured reproducibility thresholds.

[0113] For metrics measured longitudinally across multiple timepoints, a maximum CV (Max CV) may be computed as an initial reproducibility assessment. The CV is calculated for each timepoint, and the maximum value defines the reproducibility status. For instance, in the example process 300a, the server 302 first computes the CV (or other types of variability measures) for each of the timepoint metrics. The server 302 then determines the maximum CV value (Max CV) of all of the computed CV values and uses the maximum CV to determine the reproducibility status. A Max CV < 15% is a threshold value that indicates an “acceptable” reproducibility status across MPS-related samples or studies, and a Max CV < 5% is a threshold value that indicates an “excellent” reproducibility status across the MPS-related samples or studies.

Page 30 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0114] If the server 302 determines the variability or reproducibility value of the multiple timepoint metrics satisfies a corresponding minimal reproducibility threshold (e.g., CV < 15%) of an “acceptable” reproducibility status, then the server 302 proceeds to operation 317 to invoke and execute one or more downstream operations. Otherwise, the server 302 proceeds to operation 315.

[0115] In operation 315, after the server 302 determines the reproducibility value fails the corresponding threshold values, then server 302 performs one or more operations for computing secondary or other forms of variability values to evaluate and indicate, for example, the differences in both a magnitude and trends amongst the MPS models or samples. The server 302 then determines whether the additional types of variability values satisfy another reproducibility threshold score.

[0116] In the example process 300b, when the server 302 computes the Max CV for the multiple timepoint metrics and determines the Max CV fails to satisfy the Max CV threshold (e.g., Max CV > 15%), then the server 302 may compute the ICC for the multiple timepoint metrics, where the ICC assess and indicates the differences in both the magnitude and trends amongst the various MPS models or samples. An ICC >0.8 may indicate excellent reproducibility, and 0.2 < ICC < 0.8 may indicate acceptable reproducibility, and an ICC < 0.2 may indicate poor reproducibility. If the ICC reproducibility is poor, then the server normalizes each of the multiple timepoint trends to their respective median value and recalculates the ICC, which may result or output a Normalized ICC. This provides a measure of reproducibility of the trends without considering magnitude. A Normalized ICC > 0.8 may indicate excellent reproducibility, and where 0.2 < Normalized ICC < 0.8, the Normalized ICC may indicate acceptable reproducibility, and normalized ICC < 0.2 may indicate poor reproducibility.

[0117] In operation 317, the server 302 determines and generates a reproducibility status output. When the server 302 determines that the Max CV for the multiple timepoint metrics satisfied the minimally acceptable threshold reproducibility value or an excellent (or highest) threshold reproducibility (in prior operation 313), then the server 302 determines the reproducibility status is “acceptable” or “excellent.” The server 302 may further perform certain downstream operations based upon determining whether the similarity or probability score satisfied or failed the other reproducibility threshold (in prior operation 315). The server 302 stores

Page 31 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT the various types of experiment results data and reproducibility status output into one or more databases.

[0118] The server 302 executes or performs one or more downstream operations in accordance with the reproducibility status output. The server 302 may generate, for example, one or more messages for display on a graphical user interface for the clinician-user, indicating the reproducibility status.

[0119] The server 302 may use the experiment results data and other reproducibility status outputs for training or retraining one or more machine-learning models. As an example, the reproducibility status may trigger the server 302 to retrain or adjust the hyperparameters of one or more machine-learning models (e.g., MPS model configuration engine, MPS study configuration engine, cohort prediction engine) as negative feedback for an increased level of error, when the server 302 determines that the reproducibility status output is “poor” or otherwise indicates that the MPS models or MPS study are not reproducible and/or lack heterogeneity. Likewise, the reproducibility status may trigger the server 302 to retrain or adjust the hyperparameters of one or more machine-learning models (e.g., MPS model configuration engine, MPS study configuration engine, cohort prediction engine) as positive or reinforced feedback for a decreased level of error, when the server 302 determines that the reproducibility status output is “excellent,” “acceptable,” or otherwise indicates that the MPS models or MPS study are reproducible and/or heterogeneous.

[0120] FIG. 4A shows dataflow amongst software components of a server 402 (e.g., analytics server 102, server 302) of a system performing a process 400 for evaluating reproducibility and heterogeneity of MPS models across a plurality of experiments of an MPS- based study. Three levels of reproducibility can be used to determine the reproducibility of MPS models or MPS study starting with a reproducibility evaluation of an absolute signal level across MPS studies, followed by a reproducibility evaluation of a relative signal, and then performing a reproducibility evaluation of overall trends in MPS model responses across MPS studies. FIG. 4B includes graphical user interfaces 420a-420c (generally referred to as a graphical user interface 420) displaying data measurements and reproducibility status outputs generated for the evaluation process 400.

Page 32 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0121] For ease of description, the server 402 executes the functions and features of the process 400, though embodiments are not so limited. Various types of computing devices may perform the process 400. Moreover, multiple computing devices may perform the features and functions mentioned in the descriptions of FIG. 4A.

[0122] The server 402 obtains (e.g., receives, retrieves) subject data and experiment results data from one or more data sources (not shown) of the system. The data sources may include client computing devices of end-users (e.g., user device 114) or one or more databases of the system (e.g., analytics database 104, subject databases 106, conditioned databases 108), among other types of sources of subject data or experiment results data accessible to the server 402. The server 402 executes the various functions for evaluating reproducibility and heterogeneity in response to an instruction to invoke or otherwise execute the particular software functions. The server 402 may receive a request containing the instruction from a client device or input from a graphical user interface. In this way, an end-user may manually request and instruct the server 402 to evaluate the reproducibility and heterogeneity of the MPS models. Additionally or alternatively, the server 402 may receive configuration inputs that preconfigure the server 402 with the instruction to invoke the functions at a predetermined interval or in response to receiving updated data (e g., updated subject data, updated experiment results data, updated experiment configuration parameters, updated MPS model configuration parameters). In the way, the software of the server 402 may be preconfigured with the instruction to automatically invoke the functions for evaluating reproducibility and heterogeneity for the MPS models.

[0123] In operation 401, the server 402 determines a reproducibility status for absolute signal data (in the experiment results data) for the MPS models, using measurement values for one or more MPS models across one or more MPS studies. The server 402 then determines whether the absolute signal reproducibility status satisfies an absolute reproducibility threshold for the absolute signal of the MPS models for the MPS studies.

[0124] As an example, during an MPS study observing a Metabolic-Dysfunction Associated Steatotic Liver Disease (MASLD), the MPS model includes a cellular structure forming a liver or portion of liver for a subject. A measurement instrument may capture and return values for an amount of lipid droplet volume or indicating an amount of protein secreted by the MPS model at a given timepoint, as an indication of steatosis in the MASLD MPS. In this example,

Page 33 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT experiment results data includes measurement data values from five independent MPS studies for different MPS models where the lipid droplet volumes are measured the indications of steatosis in the MASLD MPS models. With reference to FIG. 4B, the graphical user interface 420a shows that a normal fasting medium (NF) and early metabolic syndrome medium (EMS) as the MPS models do not exhibit “acceptable” absolute signal reproducibility across studies using the CV variability values computed for the absolute signal level, but the MPS models do exhibit “acceptable” absolute signal reproducibility across the studies using ANOVA on the absolute signal level.

[0125] Turning back to FIG. 4A, if the server 402 determines that the absolute signal reproducibility satisfied a minimally acceptable threshold reproducibility value or an “excellent” (or highest) threshold reproducibility value and status (in current operation 401), then the server 402 may proceed directly to later operation 407. Otherwise, the server 402 proceeds to operation 403

[0126] In operation 403, the server 402 determines a reproducibility status for relative signal data for the MPS models, using one or more relative measurement values for an MPS model. The server 402 then determines whether the relative signal reproducibility status satisfies a relative reproducibility threshold.

[0127] Continuing the earlier example, the server 402 obtains the relative signal data for the MPS models for the MPS studies, where the relative signal data includes a Collagen 1A1 secretion captured by the server 402 or other measurement instruction as the measurement data value metric for early fibrosis in the MPS models (e.g., NF, EMS). With reference to FIG. 4B, the graphical user interface 420b shows that, when the server 402 normalized the EMS data to the NF data, the NF and EMS as the MPS models exhibit “excellent” reproducibility as the relative signal values across the five studies.

[0128] Turning back to FIG. 4A, if the server 402 determines that the relative signal reproducibility satisfied the “acceptable” threshold reproducibility value or the “excellent” threshold reproducibility value and status (in current operation 403), then the server 402 may proceed directly to the later operation 407. Otherwise, the server 402 proceeds to operation 405.

Page 34 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0129] In operation 405, the server 402 determines a reproducibility status for trends for the MPS models. The server 402 then determines whether the trends reproducibility status satisfies a trends reproducibility threshold.

[0130] Continuing with the example, the server 402 obtains secreted concentrations of IL-6 as the measurement values or metrics from the experiment results data and/or measurement instrument. With reference to FIG. 4B, the graphical user interface 420c shows that, when the server 402 normalizes the EMS data to the NF data and computes trends for changes in IL-6 secretion, the resulting trends indicate the NF and EMS as the MPS models exhibit “poor” reproducibility status. The graphical user interface 420c may indicate, for example, secreted concentrations of IL-6 displayed a consistent increase in the EMS condition (of the particular MPS model for the EMS condition) over the NF condition (of the particular MPS model for the EMS condition) in each of the MPS studies.

[0131] In operation 407, the server 402 determines and generates a reproducibility status output. When the server 402 determines that the absolute signal reproducibility status or the relative signal reproducibility statis satisfies an “acceptable” or “excellent” threshold reproducibility value (in prior operation 401 or operation 403), then the server 402 determines a final reproducibility status is “acceptable” or “excellent.” Alternatively, the server 402 may further perform certain downstream operations based on determining whether the trends reproducibility status satisfied or failed the trends reproducibility threshold (in prior operation 405). The server 402 stores the various types of experiment results data and reproducibility status output into one or more databases.

[0132] The server 402 executes or performs one or more downstream operations in accordance with the reproducibility status output for the one or more levels of data signals (e.g., absolute signal, relative signal, trends). The server 402 may generate, for example, one or more messages for display on a graphical user interface for the clinician-user, indicating the final reproducibility status or the reproducibility status of one or more levels. The server 402 may use the experiment results data and other reproducibility status outputs for training or retraining one or more machine-learning models.

Page 35 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0133] FIG. 5 shows the dataflow in a computer-executed process 500 for determining heterogeneity when the MPS model is reproducible. A server executes the process 500 to compute, generate, or otherwise determine the reproducibility and/or heterogeneity for the MPS models using preconfigured metrics in experiment results data. Various types of computing devices may perform the process 500. Moreover, multiple computing devices may perform the features and functions mentioned in the descriptions of FIG. 5.

[0134] In operation 501, the server determines an overall variability value amongst patient samples using the measurement metrics as indicated by configuration parameters or user instructions.

[0135] As an example, the experiment results data includes measurement data values captured during development of steatosis in MPS models for the patients’ liver cells. The individual MPS models constructed with primary hepatocytes from five different patients and tested in a MASLD disease model across several studies under normal fasting (NF) and early metabolic syndrome (EMS) conditions, and which included stellate, and Kupfer cells. The server computes an ANOVA variability value to compare a degree of steatosis induced in the MPS model across the patient cell samples and experimental conditions.

[0136] The server may then compute an overall variability among different patient cell lots with different disease states, where the overall variability value computed by the server indicated a high level or degree of variability as shown in FIG. 5.

[0137] In operation 503, the server parses or segregates the data sets based upon a state of the MPS models. Continuing with the example, the server segregates the data for the patient cells data based on a disease state (e.g., NF medium is non-disease, EMS medium is diseased). The server may compute the variability determine whether the variability, heterogeneity, or homogeneity value satisfies a heterogeneity threshold and/or homogeneity threshold. As shown in FIG. 5, the server determines there is a high degree of variability among the patient specific MPS models.

[0138] In operation 505, the server parses or segregates the data sets according additional patient attributes (e.g., genotypes), computes a variability value, and determines whether the variability, heterogeneity, or homogeneity value satisfies the same or different heterogeneity

Page 36 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT threshold and/or homogeneity threshold for the patient attribute. Continuing with the example, the server segregates the data for the patient cells data based on genotype (CC = wt, GG = risk variant). As shown in FIG. 5, the server determines that this additional segregation reduced the variability values across the MPS experiments. For instance, the variability values may indicate that the patient cells having a ‘CC’ genotype respond differently than the patient cells with the ‘GG’ genotype variant.

[0139] In operation 507, the server identifies and returns the subsets of patient datasets, attributes, or MPS models having variability values that correspond to an “excellent” reproducibility status, as well as the patient datasets, attributes, or MPS models indicating or suggestive of heterogeneity. In this way, the server identifies configuration parameters for MPS models or MPS studies that have reproducibility and heterogeneity and avoids confusing desirable biological/clinical heterogeneity with unwanted experimental variability.

[0140] FIG. 6 is a flowchart of a method 600 for analyzing cell-based in vitro data and assessing reproducibility of physical models with computing systems. For ease of description, a computer executes the functions and features of the method 600, though embodiments are not so limited. Various types of computing devices may perform the process 600. Moreover, multiple computing devices may perform the features and functions mentioned in the descriptions of FIG. 6

[0141] At operation 610, the computer obtains, from a database, experiment results data for a plurality of MPS models of a plurality of MPS studies, indicating a plurality of measurement data values at a plurality of timepoints in an MPS study.

[0142] At operation 620, the computer generates a first single timepoint variability value based upon each respective measurement data value for the MPS models at a given timepoint of the MPS study.

[0143] At operation 630, the computer determines that the first single timepoint variability value fails to satisfy a first reproducibility threshold. In response to determining that the first single timepoint variability value fails to satisfy a first reproducibility threshold, the computer generates a second single timepoint variability value based upon each respective measurement data value for the MPS models at the given timepoint in each MPS study.

Page 37 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

[0144] At operation 640, the computer generates or assigns a reproducibility status for the MPS models based upon the first single timepoint variability value and the second single timepoint variability value. In some cases, in response to the computer determining that a heterogeneity and/or homogeneity variability value satisfies a heterogeneity or homogeneity threshold to assign the reproducibility status, the computer identifies the homogeneity or heterogeneity in a certain subset of MPS models and assigns the reproducibility status or indicator. In some cases the computer in response to and based upon determining that a subset of MPS models include varied cell attributes for a plurality of subjects (e.g., patients), the computer identifies or determines the certain subset of MPS models has a non-reproducibility status.

[0145] The various illustrative logical blocks, modules, circuits, and algorithm steps described in connection with the embodiments disclosed herein may be implemented as electronic hardware, computer software, or combinations of both. To clearly illustrate this interchangeability of hardware and software, various illustrative components, blocks, modules, circuits, and steps have been described above generally in terms of their functionality. Whether such functionality is implemented as hardware or software depends upon the particular application and design constraints imposed on the overall system. Skilled artisans may implement the described functionality in varying ways for each particular application, but such implementation decisions should not be interpreted as causing a departure from the scope of the present invention.

[0146] Embodiments implemented in computer software may be implemented in software, firmware, middleware, microcode, hardware description languages, or any combination thereof. A code segment or machine-executable instructions may represent a procedure, a function, a subprogram, a program, a routine, a subroutine, a module, a software package, a class, or any combination of instructions, data structures, or program statements. A code segment may be coupled to another code segment or a hardware circuit by passing and/or receiving information, data, arguments, attributes, or memory contents. Information, arguments, attributes, data, etc. may be passed, forwarded, or transmitted via any suitable means including memory sharing, message passing, token passing, network transmission, etc.

[0147] The actual software code or specialized control hardware used to implement these systems and methods is not limiting of the invention. Thus, the operation and behavior of the systems and methods were described without reference to the specific software code being

Page 38 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT understood that software and control hardware can be designed to implement the systems and methods based on the description herein.

[0148] When implemented in software, the functions may be stored as one or more instructions or code on a non-transitory computer-readable or processor-readable storage medium. The steps of a method or algorithm disclosed herein may be embodied in a processor-executable software module which may reside on a computer-readable or processor-readable storage medium. A non-transitory computer-readable or processor-readable media includes both computer storage media and tangible storage media that facilitate transfer of a computer program from one place to another. A non-transitory processor-readable storage media may be any available media that may be accessed by a computer. By way of example, and not limitation, such non-transitory processor- readable media may comprise RAM, ROM, EEPROM, CD-ROM or other optical disk storage, magnetic disk storage or other magnetic storage devices, or any other tangible storage medium that may be used to store desired program code in the form of instructions or data structures and that may be accessed by a computer or processor. Disk and disc, as used herein, include compact disc (CD), laser disc, optical disc, digital versatile disc (DVD), floppy disk, and blu-ray disc where disks usually reproduce data magnetically, while discs reproduce data optically with lasers. Combinations of the above should also be included within the scope of computer-readable media. Additionally, the operations of a method or algorithm may reside as one or any combination or set of codes and/or instructions on a non-transitory processor-readable medium and/or computer- readable medium, which may be incorporated into a computer program product.

[0149] The preceding description of the disclosed embodiments is provided to enable any person skilled in the art to make or use the present invention. Various modifications to these embodiments will be readily apparent to those skilled in the art, and the generic principles defined herein may be applied to other embodiments without departing from the spirit or scope of the invention. Thus, the present invention is not intended to be limited to the embodiments shown herein but is to be accorded the widest scope consistent with the following claims and the principles and novel features disclosed herein.

[0150] While various aspects and embodiments have been disclosed, other aspects and embodiments are contemplated. The various aspects and embodiments disclosed are for purposes

Page 39 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT of illustration and are not intended to be limiting, with the true scope and spirit being indicated by the following claims.

Page 40 of 47

4933-5561-2507.1

Claims

076333-1039 / 06749 PATENT CLAIMS What is claimed is:

1. A computer-implemented method for analyzing in-vitro data, the method comprising: obtaining, by a computing device, experimenting results data associated with a plurality of microphysiology system constructs of a plurality of microphysiology system experimenting operations sequences, indicating a plurality of measurement data values at a plurality of interval points in a microphysiology system experimenting operations sequence; generating, by the computer, a first single interval point variability value based upon each respective measurement data value for the microphysiology system at a given interval point of the microphysiology system experimenting operations sequence; in response to determining that the first single interval point variability value does not satisfy a first reproducibility threshold: generating, by the computer, a second single interval point variability value based upon each respective measurement data value for the microphysiology system at the given interval point in each microphysiology system experimenting operations sequence; and generating, by the computer, a reproducibility status for the microphysiology system based upon the first single interval point variability value and the second single interval point variability value.

2. The method according to claim 1, further comprising: generating, by the computer, a homogeneity variability value using the experiment results data for a first subset of microphysiology system constructs having a predetermined attribute value; and identifying, by the computer, a homogeneity in the first subset of microphysiology system constructs in response to determining that the homogeneity variability value satisfies a homogeneity threshold to assign a reproducibility status.

3. The method according to claim 2, further comprising: selecting, by the computer, the first subset of microphysiology system constructs including an identical set of cell attributes corresponding to a same subject and associated with a particular experimental treatment;

Page 41 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT selecting, by the computer, a second subset of microphysiology system constructs including a plurality of varied cell attributes associated with a plurality of subjects; and identifying, by the computer, a biological heterogeneity status for the first subset of microphysiology system constructs for the same subject distinct from an experimental variability for the first subset of microphysiology system constructs, wherein the computer identifies the biological heterogeneity status in response to determining that the second subset of microphysiology system constructs including the plurality of varied cell attributes for the plurality of subjects has a non-reproducibility status.

4. The method according to claim 1, further comprising determining by the computer, whether the second single interval point variability value satisfies a second reproducibility threshold.

5. The method according to claim 1, further comprising: generating, by the computer, a first multi-interval point variability value for the microphysiology system constructs of the plurality of microphysiology system experimenting operations sequences based upon the plurality of respective measurement data values for the microphysiology system constructs at each interval point; and determining, by the computer, whether the first multi-interval point variability value for the microphysiology system constructs satisfies a first multi-interval point reproducibility threshold.

6. The method according to claim 5, further comprising: in response to determining that the first multi-interval point variability value fails to satisfy a the first multi -interval point reproducibility threshold: generating, by the computer, a second multi -interval point variability value for the microphysiology system constructs based upon the plurality of measurement data values for the plurality of microphysiology system constructs at the plurality of interval points in each microphysiology system experimenting operations sequence; and determining, by the computer, whether the second multi -interval point variability value for the microphysiology system constructs satisfies a second multi -interval point reproducibility threshold.

Page 42 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

7. The method according to claim 6, further comprising updating, by the computer, the reproducibility status for the microphysiology system constructs according to the first multiinterval point variability value and the second multi-interval point variability value.

8. The method according to claim 1, further comprising executing, by the computer, a microphysiology system configuration engine taking a set of microphysiology system experimenting operations sequence parameters as input, the microphysiology system configuration engine having a machine-learning model trained to generate a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

9. The method according to claim 8, further comprising re-training, by the computer, machine-learning model of the microphysiology system configuration engine in response to determining that the reproducibility status for at least one microphysiology system construct fails to satisfy a corresponding reproducibility status threshold.

10. The method according to claim 1, further comprising generating, by the computer, a user interface presenting a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

11. The method according to claim 1, wherein obtaining the experiment results data includes receiving, by the computer from a laboratory testing device, the experiment results data associated with one or more microphysiology system constructs.

12. A system for analyzing in-vitro data, the system comprising: a computer comprising at least one processor configured to: obtain, from a database, experiment results data for a plurality of microphysiology system constructs of a plurality of microphysiology system experimenting operations sequences, indicating a plurality of measurement data values at a plurality of interval points in a microphysiology system experimenting operations sequence; generate a first single interval point variability value based upon each respective measurement data value for the microphysiology system constructs at a given interval point of the microphysiology system experimenting operations sequence;

Page 43 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT in response to determining that the first single interval point variability value fails to satisfy a first reproducibility threshold, generate a second single interval point variability value based upon each respective measurement data value for the microphysiology system constructs at the given interval point in each microphysiology system experimenting operations sequence; and generate a reproducibility status for the microphysiology system constructs based upon the first single interval point variability value and the second single interval point variability value.

13. The system according to claim 12, wherein the computer is further configured to: generate a homogeneity variability value using the experiment results data for a first subset of microphysiology system constructs having a predetermined attribute value; and identify a homogeneity status for microphysiology system constructs in response to determining that the homogeneity variability value for the subset of microphysiology system constructs satisfies a homogeneity threshold to assign a reproducibility status.

14. The system according to claim 13, wherein the computer is further configured to: select the first subset of microphysiology system constructs including an identical set of cell attributes corresponding to a same subject and associated with a particular experimental treatment; select a second subset of microphysiology system constructs including a plurality of varied cell attributes associated with a plurality of subjects; and identify a biological heterogeneity status for the first subset of microphysiology system constructs for the same subject distinct from an experimental variability for the first subset of microphysiology system constructs, wherein the computer identifies the biological heterogeneity status in response to determining that the second subset of microphysiology system constructs including the plurality of varied cell attributes for the plurality of subjects has a non-reproducibility status.

15. The system according to claim 12, wherein the computer is further configured to determine whether the second single interval point variability value satisfies a second reproducibility threshold.

Page 44 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT

16. The system according to claim 12, wherein the computer is further configured to: generate a first multi-interval point variability value for the microphysiology system constructs of the plurality of microphysiology system experimenting operations sequences based upon the plurality of respective measurement data values for the microphysiology system constructs at each interval point; and determine whether the first multi -interval point variability value for the microphysiology system constructs satisfies a first multi-interval point reproducibility threshold.

17. The system according to claim 16, wherein the computer is further configured to: in response to determining that the first multi-interval point variability value fails to satisfy the first multi-interval point reproducibility threshold: generate a second multi-interval point variability value for the microphysiology system constructs based upon the plurality of measurement data values for the plurality of microphysiology system constructs at the plurality of interval points in each microphysiology system experimenting operations sequence; and determine whether the second multi-interval point variability value for the microphysiology system constructs satisfies a second multi-interval point reproducibility threshold.

18. The system according to claim 17, wherein the computer is further configured to update the reproducibility status for the microphysiology system constructs according to the first multiinterval point variability value and the second multi-interval point variability value.

19. The system according to claim 12, wherein the computer is further configured to execute a microphysiology system configuration engine taking a set of microphysiology system experimenting operations sequence parameters as input, the microphysiology system configuration engine having a machine-learning model trained to generate a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

20. The system according to claim 19, wherein the computer is further configured to re-train the machine-learning model of the microphysiology system configuration engine in response to

Page 45 of 47

4933-5561-2507.1 076333-1039 / 06749 PATENT determining that the reproducibility status for at least one microphysiology system construct fails to satisfy a corresponding reproducibility status threshold.

21. The system according to claim 12, wherein the computer is further configured to generate a user interface presenting a set of configuration parameters of one or more microphysiology system constructs of the microphysiology system experimenting operations sequence.

22. The system according to claim 12, wherein when obtaining the experiment results data, the computer is further configured to receive, from a laboratory testing device, the experiment results data associated with one or more microphysiology system constructs.

Page 46 of 47

4933-5561-2507.1