Computational system pathology spatial analysis platform for in situ or in vitro multiparametric cellular and subcellular imaging data

Abstract

The present invention relates to a computational system pathology spatial analysis platform, which includes: (i) a spatial heterogeneity quantification component, configured to generate a global quantification of spatial heterogeneity between cells of different phenotypes in multi-parameter cellular and subcellular imaging data; (ii) a microregion identification component, configured to identify multiple microregions for multiple tissue samples based on the global quantification, each microregion is associated with a tissue sample; and (iii) a weighted graph component, configured to construct a weighted graph for the multi-parameter cellular and subcellular imaging data, the weighted graph having multiple nodes and multiple edges, each edge being located between a pair of nodes, wherein in the weighted graph, each node is a specific one of the microregions, and the edge between each pair of microregions in the weighted graph indicates the degree of similarity between the pair of microregions.

Description

Background Art

[0001] 1. Field of the Invention

[0002] The present invention relates to digital pathology, and in particular, to a comprehensive computational system for pathology spatial analysis (CSPSA) computer platform capable of integrating, visualizing and modeling high-dimensional in situ or in vitro cellular and subcellular resolution imaging data.

[0003] 2. Related technologies

[0004] Digital pathology refers to the collection, storage, and display of histologically stained tissue samples, and has begun to gain traction in niche applications such as second-opinion telepathology, immunostain interpretation, and intraoperative telepathology. Typically, a large amount of patient data (consisting of 3 to 50 slides) is generated from a biopsy sample and visually evaluated by a pathologist under a microscope but using digital technology by viewing on a high-definition monitor. Current workflow practices are time-consuming, error-prone, and subjective due to the manual labor involved.

[0005] Cancer is a heterogeneous disease. In hematoxylin and eosin (H&E)-stained tissue images, heterogeneity is characterized by the presence of various histological structures, such as carcinoma in situ, invasive carcinoma, adipose tissue, blood vessels, and normal ducts. In addition, for many malignancies, molecular and cellular heterogeneity is a prominent feature between tumors from different patients, between different sites of neoplasia in a single patient, and within a single tumor. Intratumor heterogeneity involves phenotypically distinct clonal subpopulations of cancer cells and other cell types that constitute the tumor microenvironment (TME). These clonal subpopulations of cancer cells and other cell types include local and bone marrow-derived stromal stem and progenitor cells, subsets of immune inflammatory cells that can promote or kill tumors, cancer-associated fibroblasts, endothelial cells, and pericytes. The TME can be viewed as an evolving ecosystem in which cancer cells heterotypically interact with these other cell types and exploit available resources to proliferate and survive. Consistent with this view, the spatial relationship between cell types within the TME (i.e., spatial heterogeneity) appears to be one of the main drivers of disease progression and treatment resistance. Therefore, it is necessary to define the spatial heterogeneity within the TME to correctly diagnose specific disease subtypes and determine the best treatment options for individual patients.

[0006] To date, intratumor heterogeneity has been explored using three main approaches. The first approach is to collect core samples from specific areas of the tumor to measure population averages. Intra-sample heterogeneity is measured by analyzing multiple cores within the tumor using a variety of techniques, including whole-exome sequencing, epigenetics, proteomics, and metabolomics. The second approach involves isolating cells from tissue and performing "single-cell analysis" using the above methods, RNASeq, imaging, or flow cytometry. The third approach uses spatial resolution of optical microscopy imaging to maintain spatial context and is combined with molecular-specific markers to measure biomarkers in cells in situ. Biomarkers can identify cell type, activation state (e.g., phosphorylation of target proteins), and subcellular function. Although each of these methods provides a certain degree of effectiveness, they all have various disadvantages and limitations.

[0007] Additionally, one of the greatest challenges in assessing the clinical significance of tumor heterogeneity is the lack of advanced tools for spatial analysis of multiparametric cellular and subcellular imaging data. SUMMARY OF THE INVENTION

[0009] In one embodiment, a method for analyzing disease progression from multi-parameter cell and subcellular imaging data is provided, wherein the multi-parameter cell and subcellular imaging data is obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models. The method includes generating a global quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cell and subcellular imaging data, identifying multiple microregions of multiple tissue samples based on the global quantification, each microregion is associated with a corresponding one of the tissue samples, and constructing a weighted graph for the multi-parameter cell and subcellular imaging data. The weighted graph has multiple nodes and multiple edges, each edge is located between a pair of nodes, wherein in the weighted graph, each node is a specific one of the microregions, and the edge between each pair of microregions in the weighted graph indicates the degree of similarity between the pair of microregions.

[0010] In another embodiment, a computerized system for analyzing disease progression from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models is provided. The system includes a processing device, which includes: (i) a spatial heterogeneity quantification component configured to generate a global quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data; (ii) a micro-region identification component configured to identify multiple micro-regions for the multiple tissue samples based on the global quantification, each micro-region being associated with a corresponding one of the tissue samples; and (iii) a weighted graph component configured to construct a weighted graph for the multi-parameter cellular and subcellular imaging data, the weighted graph having multiple nodes and multiple edges, each edge being between a pair of nodes, wherein in the weighted graph, each node is a specific one of the micro-regions, and the edge between each pair of micro-regions in the weighted graph indicates the degree of similarity between the pair of micro-regions.

[0011] In yet another embodiment, a method is provided for generating a spatially indicative heterogeneous cellular communication representation from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models. The method includes generating a quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data, and identifying microregions of one of the tissue samples based on the quantification, including phenotyping the multi-parameter cellular and subcellular imaging data to identify certain different predetermined phenotypes, and constructing a communication graph for the microregions. Each phenotype is a node in the communication graph, and wherein the edge between each pair of phenotypes in the communication graph indicates the influence of one phenotype in the pair on the presence of another phenotype in the pair.

[0012] In yet another embodiment, a computerized system is provided for generating a spatially indicative heterogeneous cellular communication representation from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models. The system includes a processing device, which includes: (i) a spatial heterogeneity quantification component configured to generate a quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data, including performing cell phenotype identification on the multi-parameter cellular and subcellular imaging data to identify certain different predetermined phenotypes, (ii) a microregion identification component configured to identify a plurality of microregions of a plurality of tissue samples based on the quantification, each microregion being associated with one of the tissue samples, and (iii) a communication network component configured to construct a communication graph for a selected one of the microregions, wherein each phenotype is a node in the communication graph, and wherein an edge between each pair of phenotypes in the communication graph indicates an influence of one phenotype in the pair on the presence of another phenotype in the pair.

[0013] In yet another embodiment, a method for creating personalized medicine strategies for a specific patient is provided, wherein the specific patient is one of a plurality of patients, and wherein the plurality of patients are associated with multi-parameter cellular and subcellular imaging data of a plurality of tissue samples obtained from the plurality of patients. The method includes identifying a plurality of microregions in one of the tissue samples associated with the specific patient, wherein the plurality of microregions are based on the multi-parameter cellular and subcellular imaging data, generating a heterogeneous cellular communication network for each microregion, wherein each heterogeneous cellular communication network includes a spatially indicative heterogeneous cellular communication representation for the microregion, generating a quantification of the interdependence of the microregions in space and time based on the heterogeneous cellular communication network, and designing a medical strategy for the specific patient based on the quantification.

[0014] In yet another embodiment, a method for representing the temporal evolution of disease progression in a specific patient is provided, wherein the specific patient is one of a plurality of patients, and wherein the plurality of patients are associated with multi-parameter cellular and subcellular imaging data of a plurality of tissue samples obtained from the plurality of patients. The method comprises generating a geometric representation of a disease landscape for the specific patient, wherein the geometric representation comprises a plurality of points, wherein each point on the geometric representation: (i) describes the disease landscape of the specific patient at a specific time and is based on a specific one of the tissue samples associated with the selected patient, (ii) is based on a microregion in the specific one of the tissue samples that is based on the multi-parameter cellular and subcellular imaging data, and (iii) comprises a heterogeneous cellular communication network for the microregion, comprising a spatially indicative heterogeneous cellular communication representation for the microregion. BRIEF DESCRIPTION OF THE DRAWINGS

[0015] Figure 1 is a schematic diagram of an exemplary computational system for pathology spatial analysis (CSPSA) platform for in situ or in vitro multi-parameter cellular and subcellular imaging data according to an embodiment of the disclosed concept;

[0016] Figure 2 is a schematic representation of an exemplary supermultiple image stack;

[0017] Figure 3 is an exemplary embodiment according to the disclosed concept, showing that Figure 1 Flowchart of a method for analyzing tumor progression / evolution from multi-parameter cellular and subcellular imaging data obtained from multiple tumor sections of a patient cohort, implemented in the CSPSA platform of FIG.

[0018] Figure 4 is a schematic representation of an exemplary global space diagram;

[0019] Figure 5A schematic representation of each predetermined primary biomarker intensity pattern of an exemplary embodiment of the disclosed concept is shown;

[0020] Figure 6 shows a schematic representation of a spatially dependent image of cells according to one specific exemplary embodiment of the disclosed concept;

[0021] Figure 7 is a schematic representation of an exemplary microregion in an exemplary tissue section according to an exemplary embodiment of the disclosed concepts;

[0022] Figure 8 is a schematic representation of a weighted micro-region map constructed for multi-parameter cellular and subcellular imaging data according to an exemplary embodiment of the disclosed concepts;

[0023] Fig. 9 is a flow chart illustrating a method for generating a spatially indicative heterogeneous cellular communication representation from multi-parameter cellular and subcellular imaging data according to another exemplary embodiment of the disclosed concept;

[0024] Fig.10 is a schematic representation of an exemplary communication diagram according to an exemplary embodiment of the disclosed concepts;

[0025] Fig.11 is a schematic diagram of a particular communication graph generated according to a particular exemplary embodiment, wherein the phenotypes are tumor cell, lymphocyte, macrophage, stroma, and necrosis;

[0026] Fig.12 is a schematic representation of a personalized medicine strategy according to one aspect of the disclosed concept; and

[0027] Fig.13 is a schematic representation of an exemplary cancer landscape according to an aspect of the disclosed concept. DETAILED DESCRIPTION

[0028] As used herein, the singular forms "a," "an," and "the" include plural references unless the context clearly dictates otherwise.

[0029] As used herein, the statement that two or more parts or components are “coupled” shall mean that the parts are joined or function together directly or indirectly (ie, through one or more intermediate parts or components, so long as a connection exists).

[0030] As used herein, the term "number" shall mean one or an integer greater than one (ie, a plurality).

[0031] As used herein, the terms "component" and "system" are intended to refer to a computer-related entity, hardware, a combination of hardware and software, software, or software in execution. For example, a component can be, but is not limited to, a process running on a processor, a processor, an object, an executable file, an execution thread, a program, and / or a computer. As an illustration, both an application running on a server and a server can be a component. One or more components can reside in a process and / or execution thread, and a component can be located on a computer and / or distributed between two or more computers. Although certain ways of presenting information to a user are shown and described with respect to certain figures or graphics (as screen shots), those skilled in the relevant art will recognize that various other alternatives can be used.

[0032] As used herein, the term "multiparametric cellular and subcellular imaging data" refers to data obtained by generating a large number of images from multiple tissue sections, which provides information on multiple measurable parameters at the cellular or subcellular level of the tissue sections. Multiparametric cellular and subcellular imaging data can be created by a variety of different imaging techniques, such as, but not limited to (1) transmitted light imaging with IHC using multiple biomarkers, (2) immunofluorescence imaging including multiplex imaging (1-7 biomarkers) and super-multiplex imaging (>7 biomarkers), (3) toponome imaging, (4) matrix-assisted laser desorption / ionization mass spectrometric imaging (MALDI MSI), (5) complementary spatial imaging, such as FISH, MxFISH, FISHSEQ or CyTOF, (6) multiparametric ion beam imaging, and (7) in vitro imaging of experimental models. Additionally, for example and without limitation, multi-parameter cellular and subcellular imaging data may be created by labeling each tissue section with multiple different biomarkers and generating multiple biomarker images from multiple tissue sections.

[0033] As used herein, the term "spatial map" shall refer to a collection of data of multiple quantified spatial statistics and / or a visually perceptible form of representation that indicates the relationship between cells of different predetermined phenotypes in a set of multi-parameter cellular and subcellular imaging data. For example, but not limited to, the spatial map can be a point mutual information (PMI) map generated in the manner described in PCT Application No. PCT / US2016 / 036825 and U.S. Patent Application Publication No. 2018 / 0204085, both of which are entitled "Systems and Methods for Finding Regions of Interest in Hematoxylin and Eosin (H&E) Stained Tissue Images and Quantifying Intratumor Cellular Spatial Heterogeneity in Multiplexed / Hyperplexed Fluorescence Tissue Images", the disclosures of which are incorporated herein by reference.

[0034] As used herein, the term "microdomain" refers to a phenotypically distinct spatial arrangement (or motif) of cells in a tissue sample that results from spatial intratumor heterogeneity and is associated with one or more outcome-specific variables (e.g., time to recurrence). Microdomains can be identified from multiparameter cellular and subcellular imaging data according to any of a number of known or later developed methods, such as, but not limited to, those described in U.S. Provisional Application Serial No. 62 / 675,832 (invention title: "Predicting the Recurrence Risk of Cancer Patients From Primary Tumors with Multiplexed Immunofluorescence Biomarkers and Their Spatial Correlation Statistics"), and PCT Application No. PCT / US2019 / 033662 (invention title: "System and Method for Predicting the Risk of Cancer Recurrence From Spatial Multi-Parameter Cellular and Sub-Cellular Imaging Data for Tumors by Identifying Emergent Spatial Domain Networks Associated With Recurrence"), each of which is incorporated herein by reference, and / or in Spagnolo et al., Platform for Quantitative Evaluation of Spatial Intratumoral Heterogeneity in Multiplexed Fluorescence Images, Cancer Res. 2017 Nov. 1;77(21):e71-e74 and / or implemented in the public domain THRIVE (Tumor Heterogeneity Research Interactive Visualization Environment) software described in the above-mentioned reference by Spagnolo et al. In this method, spatially resolved correlations between biomarkers are used as covariates for outcome data (e.g., recurrence) in multivariate survival models to create maps of the spatial organization of cancer recurrence in multiple tissue samples. These maps depict microregions associated with recurrence and metastatic progression.

[0035] Directional phrases used herein, such as, but not limited to, top, bottom, left, right, upper, lower, front, back and their derivatives, relate to the orientation of elements shown in the drawings and are not limited to the claims unless explicitly stated.

[0036] For the purpose of explanation, the concepts disclosed in the present invention will now be described in conjunction with many specific details to provide a thorough understanding of the present invention. However, it will be apparent that the concepts disclosed in the present invention can be practiced without these specific details without departing from the spirit and scope of the present invention.

[0037] The concepts disclosed in the present invention provide a comprehensive computational systems pathology spatial analysis (CSPSA) platform that can integrate, visualize and / or model high-dimensional in situ cell and subcellular resolution imaging data. The CSPSA platform of the concepts disclosed in the present invention combines a core set of existing cell phenotype identification and spatial analysis tools, as well as a set of advanced tool sets for inferring spatial heterogeneous cell communication (cell to cell) and intracellular communication patterns, and establishing tumor (or other disease) evolutionary trees from precancerous origins to metastatic endpoints (in the case of cancer). Combined together, these tools can be used for both research and clinical purposes, such as quantifying spatial intratumor heterogeneity in tissue samples and in vitro models, and associating them with outcome data, establishing diagnostic and prognostic applications, and designing personalized treatment strategies and drug discovery. In addition, the platform can be strengthened by enriching in situ cell and subcellular resolution images with regional specific genomics and spatial transcriptomics data.

[0038] Figure 1 is a schematic diagram of an exemplary computational system for pathology spatial analysis (CSPSA) platform 5 for in situ multi-parameter cellular and subcellular imaging data according to an embodiment of the presently disclosed concepts, in which the various methodologies described herein may be implemented. Figure 1 As shown, the CSPSA platform 5 is a computing device that is constructed to receive and store certain multi-parameter cellular and subcellular imaging data 10 (e.g., a large patient cohort of tumors including recurrent and non-recurrent cancers, or a single patient or a single tumor) and process such data 10 as described herein. In a non-limiting exemplary embodiment, multiplex or super-multiplex immunofluorescence imaging is used to generate the multi-parameter cellular and subcellular imaging data 10, but it should be understood that other imaging techniques, such as those described elsewhere herein, may also be used. For example, the multi-parameter cellular and subcellular imaging data 10 may be based on an exemplary super-multiplex image stack 12 of all patient data in a cohort (e.g., Figure 2 shown).

[0039] The CSPSA platform 5 can be, for example, but not limited to, a PC, a notebook computer, a tablet computer, a smart phone, or any other suitable computing device constructed to perform the functions described herein. The CSPSA platform 5 includes an input device 15 (e.g., a keyboard), a display 20 (e.g., an LCD), and a processing device 25. As described in detail herein, a user can use the input device 15 to provide input to the processing device 25, and the processing device 25 provides an output signal to the display 20 so that the display 20 can display information (such as a spatial map or other spatially dependent image, a micro-area image, a weighted map image, and / or a communication network map image) to the user. The processing device 25 includes a processor and a memory. The processor can be, for example, but not limited to, a microprocessor (μP), a microcontroller, an application specific integrated circuit (ASIC), or some other suitable processing device, which interfaces with the memory. The memory can be any one or more of various types of internal and / or external storage media, such as, but not limited to, RAM, ROM, EPROM, EEPROM, FLASH, etc., which provide storage registers, i.e., machine-readable media, for data storage such as in the form of an internal storage area of ​​a computer, and can be a volatile memory or a non-volatile memory. The memory stores a number of routines executable by the processor, including routines for implementing the concepts disclosed herein. In particular, the processing device 25 includes a spatial heterogeneity quantification component 30 configured to generate a global quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data 10 as described herein, a micro-region identification component 35 configured to identify a plurality of micro-regions for a plurality of tumor slices from the multi-parameter cellular and subcellular imaging data 10 (as described herein), a weighted graph component 40 configured to construct a weighted graph for the multi-parameter cellular and subcellular imaging data 10 (as described herein) based on the global quantification generated by the spatial heterogeneity quantification component 30, and a communication network component 45 configured to construct a communication network graph for the multi-parameter cellular and subcellular imaging data 10 (as described herein).

[0040] Figure 3 is a flow chart illustrating a method for analyzing tumor progression / evolution from multi-parameter cellular and subcellular imaging data 10 obtained from multiple tumor sections of a patient cohort, which may be implemented in a CSPSA platform 5 according to an exemplary embodiment of the disclosed concepts. However, it will be understood that this is merely exemplary and Figure 3 The steps of the methods shown in may be implemented in other configurations and / or platforms.

[0041] The method begins at step 50, where the spatial heterogeneity quantification component 30 generates a global quantification of spatial heterogeneity in the multi-parameter cellular and subcellular imaging data 10. In an exemplary embodiment, the global quantification generated in step 50 is a global spatial map for the multi-parameter cellular and subcellular imaging data 10. In a specific embodiment, the global spatial map is a global PMI map 52 (e.g., Figure 4 ), which is generated in the manner described in PCT Application No. PCT / US2016 / 036825 and U.S. Patent Application Publication No. 2018 / 0204085, which are mentioned elsewhere in this article and incorporated herein by reference. In this exemplary embodiment, multiple biomarker images are generated from multiple tissue sections by labeling each tissue section with multiple different biomarkers (e.g., ER, PR, and HER2) to create multi-parameter cell and subcellular imaging data 10. As discussed in the aforementioned application, a PMI map 52 is generated by first performing cell segmentation on the multi-parameter cell and subcellular imaging data 10 (i.e., on each "sheet" thereof). Any of many known or later developed suitable cell segmentation algorithms can be used. Then, the spatial position and biomarker intensity data of each cell are obtained, and each cell is assigned to one of the predetermined phenotypes (each phenotype is a predetermined main biomarker intensity pattern) based on the biomarker intensity composition of the cell. Figure 5 Schematic representations 56 (labeled 1 to 8) of each predetermined primary biomarker intensity pattern of the exemplary embodiment are shown. In the exemplary embodiment, each schematic representation 56 is provided in a unique color or colors so that the schematic representations can be easily distinguished from each other. Figure 5 The schematic representation shown in can be used to generate cell spatial dependence images that visually demonstrate the heterogeneity of subject tissue samples. Figure 6 A cell space dependency image 58 is shown according to one specific exemplary embodiment of the disclosed concept. Figure 6 As shown, the cell spatial dependence image 58 shows the spatial dependence between cells of the subject slide (using the schematic representation 56). Next, a spatial network was constructed to describe the organization of the main biomarker intensity patterns in the subject slide. The heterogeneity of the subject slide was then quantified by generating a PMI map 52 (as shown in FIG. Figure 4 In an exemplary embodiment, the spatial network and PMI map 52 are generated as follows.

[0042] In order to represent the spatial organization of biomarker patterns in the biomarker images (i.e., tissue / tumor samples) of the subject's map, a network was constructed for the subject's map. The construction of the spatial network for tumor samples essentially couples the cell biomarker intensity data (in the nodes of the network) to the spatial data (in the edges of the network). The assumption in the network construction is that the cell has the ability to communicate with nearby cells within certain limits (e.g., up to 250 μm), and the ability of the cell to communicate within the limit depends on the distance of the cell. Therefore, the probability distribution in the exemplary embodiment is calculated for the distance between the cell in the subject's map and its 10 nearest neighbors. A hard limit is selected according to the median of this distribution multiplied by 1.5 (to estimate the standard deviation), where the cells in the network are connected only within this limit. Then, the edges between the cells in the network are weighted by the distance between adjacent cells.

[0043] Next, in an exemplary embodiment, point mutual information (PMI) is used to measure the association between each pair of biomarker patterns in the dictionary in the subject's image, thereby measuring the association between different cell phenotypes. This index captures the general statistical association of linear and nonlinear, where previous studies used linear indicators, such as Spearman's rho coefficient. Once the PMI is calculated for each pair of biomarker patterns, the metrics of all associations in the subject's image data are displayed in the PMI graph 52. It will be understood that the use of PMI in this embodiment is merely exemplary and that other methods of defining spatial relationships, such as but not limited to Spagnolo DM, Al-Kofahi Y, Zhu P, Lezon TR, Gough A, Stern AM, Lee AV, Ginty F, Sarachan B, Taylor DL, Chennubhotla SC, Platform for Quantitative Evaluation of Spatial Intratumoral Heterogeneity in Multiplexed Fluorescence Images, Cancer Res. 2017 Nov 1, and Nguyen, L., Tosun, B., Fine, J., Lee, A., Taylor, L., Chennubhotla, C. (2017), Spatial statistics for segmenting histological structures in H&E stained tissue images, IEEE Trans Med Imaging. 2017 Mar 16 may be employed in conjunction with the disclosed concepts.

[0044] Exemplary PMI Figure 52 describes the relationship between different cell phenotypes in the microenvironment of the subject's figure. In particular, the entry in PMI Figure 52 indicates the frequency of a specific spatial interaction between two phenotypes (with reference to row and column numbers) occurring in a data set (when compared to the interaction predicted by the random (or background) distribution relative to all phenotypes). The entry of the first color (e.g., red) represents a strong spatial association between the phenotypes, while the entry of the second color (e.g., black) represents a lack of any co-localization (weak spatial association between phenotypes). Other colors can be used to represent other associations. For example, the PMI entry of the third color (e.g., green) represents an association that is not stronger than the random distribution of cell phenotypes in the entire data set. In addition, PMI Figure 52 can depict anti-associations (e.g., if phenotype 1 rarely occurs near phenotype 3 in space) with an entry represented by a fourth color (e.g., blue).

[0045] Reference again Figure 3 , after step 50, the method then proceeds to step 55. At step 55, based on the global quantification generated in step 50, the microregion identification component 35 identifies and localizes one or more microregions in each tumor slice in the multi-parameter cellular and subcellular imaging data 10 (e.g., as described elsewhere herein). For example, a cancer cell type and a particular immune cell type that co-localizes with it may have a high spatial co-occurrence value. Using cancer-immune cell combinations as seed points and mining the global quantification (e.g., spatial map) to propagate phenotypic associations, in any given tissue slice, microregions can be generated by growing a spatial network of cells around the seeds based on a cutoff distance (e.g., ~100 cells). Figure 7 Such exemplary micro-domains 62 are shown in the exemplary tissue section 64 shown.

[0046] Next, at step 60, for each microregion identified in step 55, a local quantification of spatial heterogeneity (e.g., a local spatial map, such as a local PMI map) is determined. Furthermore, each determined local quantification in each tissue section is used to define the degree of similarity between pairs of microregions in the tissue section. In particular, in an exemplary embodiment, given two microregions A and B, a distance function is defined as the difference between (i) the relative abundance of cells in microregions A and B and (ii) their local quantification (e.g., a local spatial map, such as a PMI map). A and PMI B , although other methods may also be used as described herein). This process is repeated for each pair of microregions in all multi-parameter cellular and subcellular imaging data 10. This process will result in a weighted similarity being determined for each pair of microregions in each tissue section.

[0047] Next, at step 65, a method is constructed for constructing the multi-parameter cell and subcellular imaging data 10. Figure 86. Each node 68 in the weighted micro-region graph 66 is one of the micro-regions generated in step 55, and the edges 72 connecting each pair of micro-regions are weighted by their determined similarities (e.g., weighted similarity values ​​as described above). The weighted micro-region graph 66 may then be displayed on the display 20.

[0048] Then, if Figure 3 As shown in step 70 of , the weighted micro-region map 66 can be further analyzed to understand the progression of the phenotype during the metastasis process. It is worth noting that each whole tissue section in the multi-parameter cell and subcellular imaging data 10 (from relapse (R) or non-relapse (NR) patients) has a ground truth label, but there is no prior knowledge to determine which micro-region can best distinguish between the two categories. In an exemplary embodiment, labels R and NR are transferred from the tissue section to the corresponding micro-region. Then some iterations of belief propagation are performed to improve the labels, and then the micro-region groups (unified as R or NR) are isolated on the weighted micro-region map 66. By improving the confidence of the label assignment, the betweenness centrality measure can be used to identify the micro-regions that are most frequently traversed on the path from NR to R. In order to simulate the evolutionary trajectory, the weighted micro-region map 66 can be randomly walked, because it is known that some micro-regions can only be found in observed or unobserved metastasis data. The relative probabilities of the paths from NR to R can be compared to identify the event categories that are important in metastasis. The average first pass time of the random walk can then be used to identify microregions that are prone to metastasis as sets of nodes from which the probability of a random walk to R or NR is 0.5. Note that microregions with metastatic potential are the midpoints of the evolutionary trajectory from NR to R. In an exemplary embodiment, the aforementioned steps will be performed by components of the CSPSA platform 5 and displayed to the user on the display screen 20 of the CSPSA platform 5. Thus, this aspect of the disclosed concept provides the ability to define, identify, and compare phenotypes ("phenotypic evolution") that are important for metastasis, as described herein. In addition, it will be appreciated that cancer is merely an example field to which the disclosed concept can be applied, and the disclosed concept can also be applied to other disease tissues, such as neurodegenerative diseases, metabolic diseases, and inflammatory diseases, among others.

[0049] Fig. 9 is a flow chart illustrating a method of generating a spatially indicative heterogeneous cellular communication representation from multi-parameter cellular and subcellular imaging data 10, which in this embodiment is obtained from multiple tumor slices of a patient, which can be implemented in a CSPSA platform according to another exemplary embodiment of the disclosed concept. However, it will be understood that this is merely exemplary and Fig. 9 The steps of the methods shown in may be implemented in other configurations and / or platforms.

[0050] The method begins at step 75, where the spatial heterogeneity quantification component 30 generates quantification of spatial heterogeneity in the multi-parameter cell and subcellular imaging data 10, as described in detail elsewhere herein. As noted elsewhere herein, the quantification of spatial heterogeneity performed in step 75 includes performing cell phenotype identification on the multi-parameter cell and subcellular imaging data 10 to identify certain different predetermined phenotypes. In an exemplary embodiment, the quantification generated in step 75 is a spatial map for the multi-parameter cell and subcellular imaging data 10. In a specific embodiment, a spatial map is generated in a manner described in PCT Application No. PCT / US2016 / 036825 and U.S. Patent Application Publication No. 2018 / 0204085, which are mentioned elsewhere herein and incorporated herein by reference. In this exemplary embodiment, multi-parameter cell and subcellular imaging data 10 is created by generating multiple biomarker images from multiple tissue sections by labeling each tissue section with multiple different biomarkers (e.g., ER, PR, and HER2). In a non-limiting exemplary embodiment, multiplex or hyper-multiplex immunofluorescence imaging is used to generate multi-parameter cellular and subcellular imaging data 10, although it will be appreciated that other imaging techniques, such as those described elsewhere herein, may also be used.

[0051] Next, at step 80, the microregion identification component 35 identifies and locates one or more microregions (as described herein) in each tumor slice in the multi-parameter cellular and subcellular imaging data 10 based on the quantification generated at step 75. Then, at step 85, one or more identified microregions are selected. The method then proceeds to step 90.

[0052] At step 90, the communication network component 45 constructs a communication map for each selected micro-zone. The communication map may be displayed on the display 20. Fig.10 An exemplary communication diagram 92 is shown in FIG. Fig.10As shown, in an exemplary embodiment, each phenotype is a node 94 in a communication graph 92, and an edge 96 between each pair of phenotypes (each pair of nodes 94) in the communication graph 92 indicates the effect of one phenotype in the pair on the presence of the other phenotype in the pair. In this exemplary embodiment, each phenotype (each node 94) in the communication graph 92 is represented by a data vector obtained from the multi-parameter cellular and subcellular imaging data 10. In addition, for each pair of phenotypes, a numerical relationship between the pair of phenotypes is established by determining a linear or nonlinear correlation coefficient value between the data vectors of the pair of phenotypes (after removing the confounding effects of phenotypes other than the phenotypes in the pair), generating the edge 96 in the communication graph 92 in the exemplary embodiment. In addition, a directionality is established for each numerical relationship, preferably based on a plurality of receptor-ligand databases for related biomarkers. Therefore, in this embodiment, the edge 96 between each pair of nodes 94 indicates the determined numerical relationship and the determined directionality. In an exemplary embodiment, the edge 96 can be colored to represent certain correlation coefficient values ​​(e.g., positive or negative correlation).

[0053] Additionally, in an exemplary embodiment, each data vector for each node 94 is a vector of expression values ​​for a plurality of predetermined biomarkers, wherein the predetermined biomarkers are specific biomarkers used to generate the multi-parameter cellular and subcellular imaging data 10. For example, each expression value vector may be a biomarker intensity pattern for a predetermined biomarker determined in step 75. Each data vector may also be analyzed and biologically interpreted to determine a plurality of activation states 98 for each phenotype (each node 94), the activation states 98 being included in the communication graph 92. For example, the biomarker ALDH1 is a surrogate for the stemness of tumor cells, and the biomarker PD-L1 determines the mutation burden.

[0054] Fig.11 is a schematic diagram of a particular communication graph 92 generated according to a particular exemplary embodiment, wherein the phenotypes (nodes 94) are tumor cells, lymphocytes, macrophages, stroma, and necrosis. Various edge influences 96 and activation states 98 are also shown.

[0055] One result of the graphs described herein is to infer a pathway for disease progression based on the local tumor microenvironment (TME), and then list the known molecular targets in that pathway. Afterwards, machine learning tools (e.g., Balestra Web) can be used to predict drug-target interactions based on spatial relationships. This can lead to new uses for drugs, as well as new therapeutic approaches. In addition, a number of exemplary specific clinical applications are described in detail below that can incorporate various aspects of the disclosed concepts (e.g., applications of the CSPA platform 5), including: (1) drug discovery and personalized medicine strategies (using spatially modulated computational systems pathology), (2) cancer / disease landscapes, which are geometric representations of multi-parameter readouts of a patient's cancer / disease state, and (3) the CSPA platform for in vitro models for basic research and clinical translation.

[0056] In a specific exemplary embodiment, the disclosed concepts can be used for drug discovery and personalized medicine strategies (computational systems pathology with spatial modulation). More specifically, spatial intra-tumor heterogeneity is a key aspect in determining the temporal evolution of cancer and the fate of patients. This heterogeneity is reflected in the diversity of heterogeneous cellular communication networks embedded in microregions, and the resulting systems biology is therefore patient-dependent. Current treatment strategies are designed for universal patients and therefore do not reflect patient-specific systems biology. Patient-specific systems pathology begins with the identification of target microregions / regions, the characterization of the underlying communication networks, and the accurate quantification of the interdependencies of microregions in space and time. This knowledge is also necessary for new drug design strategies. This strategy can also be enhanced by using region-specific genomics.

[0057] Fig.12 is a schematic representation of a personalized medicine strategy according to this aspect of the disclosed concept, extending from tissue sections to personalized medication treatment. Fig.12 Shown from left to right, i) multi-parameter images at cellular / subcellular resolution, ii) identification of target regions (microdomains) in the images, iii) reconstruction of intra-microdomain communication networks with specific connection weights and spatial relationships between cells / microdomains, iv) combining information to extract precise systems biology, and v) defining personalized treatment strategies.

[0058] like Fig.12As shown, in the exemplary embodiment shown, for individual patients, sample microregions 1 and 2 are found to be phenotypically different. For this tissue sample, the cell phenotype identification algorithm described herein is applied. Next, as described herein, a heterogeneous cell network is constructed for each microregion. Prognostic tests performed on microregions 1 and 2, respectively, show that the risk of recurrence in both microregions is low. However, when those microregions are very close, the risk of recurrence is greatly increased, indicating that there is information flow between microregions 1 and 2. Therefore, in this aspect of the disclosed concept, the heterogeneous cell network indicates which pathways are activated and what the relationship is between the two networks. For example, WNT signaling is upregulated in microregion 2 and downregulated in microregion 1, and TGFβ is upregulated in microregion 1 and downregulated in microregion 2. In this aspect of the disclosed concept, pathways and microregion networks are identified, and therefore the information flow that must be inhibited to slow down the progression of cancer in a particular patient is also identified. With this information, personalized medical strategies can be designed.

[0059] In another specific exemplary embodiment, the disclosed concepts can be used to accurately describe the time evolution of a disease such as cancer in a particular patient. Thus, the disclosed concepts define a cancer landscape, which is a geometric representation of a multi-parameter readout of a patient's cancer state. Each point in the cancer landscape represents a patient's cancer state, such as precancerous, early, invasive, etc. The path followed in the landscape describes the time evolution of the disease in that particular patient. Current treatment practice uses a cancer landscape that is averaged across the entire patient population. Unless individual patient data exactly matches the average data, predictions made using an average model may be inaccurate.

[0060] Fig.13 An exemplary cancer landscape including a geometric representation is shown, wherein locations on the landscape represent the condition of a particular exemplary patient. Fig.13 Take colon cancer progression as an example, where the basin with label 1 represents the precancerous state, the basin with label 2 represents the small polyp state, the basin with label 3 represents the large polyp state, and the basin with label 4 represents the invasive colon cancer state. The arrows in the geometric representation show the path that the patient follows from precancerous to invasive cancer. From this model, the dynamics of cancer evolution can be inferred.

[0061] According to one aspect of this embodiment of the disclosed concept, a traceable patient cohort is tracked to construct a comprehensive library of microregions as described herein for tissue progression from precancer to metastasis. A specific heterogeneous cell communication network as described herein is associated with each microregion. In addition, each heterogeneous cell communication network has a systems biology model in the form of a system of ordinary differential equations. The system of ordinary differential equations defines the dynamics of the cancer landscape. For prospective studies, dynamics helps predict the time evolution of microregions from precancer to metastasis. In addition, the kinetic model is a proxy for the synthetic tissue generation model.

[0062] In yet another specific exemplary embodiment, the disclosed concepts can be used in conjunction with in vitro models for basic research and clinical transformation. More specifically, in two specific applications of the disclosed concepts as described above, the platform of the disclosed concepts is applied to in situ ultra-multiple single-cell resolution solid tumor imaging. In this embodiment, the same platform is applied to image data from in vitro microphysiological models. Multicellular in vitro models allow the study of spatiotemporal cell heterogeneity and heterogeneous cell communication, which recapitulates human tissues, which can be used to study the mechanisms of in vitro disease progression, test drugs, and characterize the structural organization and content of these models for potential applications in transplantation.

[0063] In vitro models come in the form of 2D cultures, 3D spheroids, organoids, and biomimetic microphysiological systems. The goal of these systems is to get closer to physiologically relevant biomimicry. 2D cultures are cell cultures grown on an adherent surface. 2D communication networks are developed, so information exchange is limited to two dimensions. 3D spheroids are clusters of cells grown in space. Although the cells are grown in an artificial environment, they better mimic in vivo growth conditions because the cells are allowed to interact and grow in all directions. Organoids are very small but self-organizing three-dimensional tissue cultures. They are usually derived from stem cells. Organoids can be made to replicate many of the complexities of an organ, or they can be directed to express selected aspects of it, such as producing only certain types of cells. The regeneration of organ function, even if partial, means the regeneration of systems biology and heterogeneous cell communication under controlled conditions in vitro. In terms of organoids, biomimetic microphysiological systems are a step forward in simulating organ function. 3D microfluidic channel cell culture chips simulate all the activities, mechanics, and physiological responses of an entire organ. This allows for a more accurate representation of systems biology and heterogeneous cell communication networks.

[0064] In one or more aspects of this particular embodiment, spatiotemporal cell heterogeneity and heterogeneous cell communication can be monitored by establishing a timeline of images and comparing the evolution of the inferred network in the model. The complexity of the in vitro system from 2D culture to biomimetic MPS is reflected in the complexity of the system biology model that can be developed. Although applicable to all possible in vitro models, the biomimetic model constructed by layering or bioprinting different cell types in appropriate 3D tissues will benefit the most by defining the spatial relationship within the model. The same type of system pathology defined in the two specific applications of the disclosed concept described above will be effective here. The purpose is to prove that the in vitro model reflects the organization and spatial relationship of the in situ tissue / organ determined in the study, and that system biology can be generalized in vitro. Models can be established and studied under "normal" healthy conditions, or cells from disease patients and / or induced pluripotent stem cells (iPSC) from patients (which have been differentiated and matured into the cell types required in the model) are established as disease models. After different time points and selected treatments, the model is fixed, embedded, sliced ​​and labeled, just like tissue from patients. Then hyperplexed imaging modalities are applied to labeled tissue slices. The computational and systems pathology analysis of the disclosed concepts described herein can then be applied to tissue sections obtained from the model.

[0065] Finally, although the disclosed concepts have been described in conjunction with imaging data obtained from tumor slices, it should be understood that the disclosed concepts may also be applied to imaging data obtained from other types of tissue slices and / or unsectioned tissue samples (using imaging modalities that can penetrate solid, unsectioned samples).

[0066] In the claims, any reference signs between brackets shall not be construed as limiting the claim. The words "comprise" or "comprising" do not exclude the presence of elements or steps other than those listed in the claim. In a device claim enumerating several means, several of these means may be embodied by the same hardware. The word "a" or "an" preceding an element does not exclude the presence of a plurality of such elements. In any device claim enumerating several means, several of these means may be embodied by the same hardware. The fact that certain elements are recited in mutually different dependent claims does not indicate that these elements cannot be used in combination.

[0067] Although the present invention has been described in detail for illustrative purposes based on what are currently considered to be the most practical and preferred embodiments, it should be understood that such details are for illustrative purposes only, and the present invention is not limited to the disclosed embodiments, but on the contrary, is intended to cover modifications and equivalent arrangements within the spirit and scope of the appended claims. For example, it should be understood that the present invention contemplates that one or more features of any embodiment may be combined with one or more features of any other embodiment to the extent possible.

Claims

1. A method for analyzing disease progression from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models, the method comprising: Generate global quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in multi-parameter cellular and subcellular imaging data; Based on global quantification, a plurality of microregions are identified for a plurality of tissue sections from multi-parameter cellular and subcellular imaging data, each microregion is associated with a corresponding one of the tissue sections, and the microregions are phenotypically distinct spatial arrangements of phenotypic cells in a tissue sample that are generated by spatial heterogeneity and are associated with one or more disease outcome-specific variables; and A weighted graph is constructed for multi-parameter cellular and subcellular imaging data, the weighted graph indicating the progression of a predetermined phenotype and having a plurality of nodes and a plurality of edges, each edge being between a pair of nodes, wherein in the weighted graph, each node is a specific one of the microregions, and the edge between each pair of microregions in the weighted graph indicates the degree of similarity between the pair of microregions, whereby the weighted graph can be analyzed to understand disease progression.

2. The method according to claim 1, further comprising determining, for each pair of micro-regions in the plurality of micro-regions, a weighted similarity between the pair of micro-regions, wherein the edge between each pair of micro-regions in the weighted graph is the weighted similarity determined for the pair of micro-regions.

3. The method of claim 2, further comprising determining, for each identified microregion, a local quantification of spatial heterogeneity between cells, wherein determining the weighted similarity between each pair of microregions is based on the local quantification of each microregion in the pair of microregions.

4. The method of claim 1, wherein the global quantification is a spatial map of multi-parameter cellular and subcellular imaging data. The method of claim 4 , wherein the spatial map is a point mutual information (PMI) map.

6. The method of claim 1, wherein the multi-parameter cellular and subcellular imaging data comprises multiplex or super-multiplex immunofluorescence imaging data.

7. The method of claim 1, wherein the predetermined phenotype comprises a predetermined set of major biomarker intensity patterns.

8. The method of claim 1, wherein a plurality of the identified microregions are associated with an outcome-specific variable in the form of time to relapse or indicative of disease progression.

9. The method of claim 1, wherein a plurality of the identified microregions are associated with an outcome-specific variable indicative of disease progression.

10. The method of claim 9, wherein the tissue sample is a tumor sample, and wherein the outcome-specific variable is indicative of metastatic potential.

11. The method of claim 1, wherein the plurality of multicellular in vitro models is selected from 2D cultures, 3D spheroids, organoids, and biomimetic microphysiological systems.

12. A non-transitory computer-readable medium storing one or more programs including instructions which, when executed by a computer, cause the computer to perform the method according to claim 1.

13. A computer system for analyzing disease progression from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multi-cellular in vitro models, the computer system comprising: A processing device, wherein the processing device comprises: a spatial heterogeneity quantification component configured to generate a global quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in multi-parameter cellular and subcellular imaging data; a microregion identification component configured to identify, for a plurality of tissue sections from the multi-parameter cellular and subcellular imaging data, a plurality of microregions based on the global quantification, each microregion being associated with a respective one of the tissue sections, and the microregions being phenotypically distinct spatial arrangements of phenotypic cells in a tissue sample that result from spatial heterogeneity and are associated with one or more disease outcome-specific variables; and A weighted graph component is configured to construct a weighted graph for multi-parameter cellular and subcellular imaging data, the weighted graph indicating the progression of a predetermined phenotype and having a plurality of nodes and a plurality of edges, each edge being located between a pair of nodes, wherein in the weighted graph, each node is a specific one of the microregions, and the edge between each pair of microregions in the weighted graph indicates a degree of similarity between the pair of microregions, whereby the weighted graph can be analyzed to understand disease progression.

14. The system of claim 13, wherein the weighted graph component is further configured to determine, for each pair of micro-regions in the plurality of micro-regions, a weighted similarity between the pair of micro-regions, wherein an edge between each pair of micro-regions in the weighted graph is the weighted similarity determined for the pair of micro-regions.

15. The system of claim 14, wherein the weighted graph component is further configured to determine, for each identified microregion, a local quantification of spatial heterogeneity between cells, wherein determining the weighted similarity between each pair of microregions is based on the local quantification of each microregion in the pair of microregions.

16. The system of claim 13, wherein the global quantification is a spatial map of the multi-parameter cellular and subcellular imaging data.

17. The system of claim 16, wherein the spatial map is a point mutual information (PMI) map.

18. The system of claim 13, wherein the multi-parameter cellular and subcellular imaging data comprises multiplexed or hyper-multiplexed immunofluorescence imaging data.

19. The system of claim 13, wherein the predetermined phenotype comprises a predetermined set of major biomarker intensity patterns.

20. The system of claim 13, wherein a plurality of the identified microregions are associated with an outcome-specific variable in the form of time to relapse or indicative of disease progression.

21. The system of claim 13, wherein a plurality of the identified microregions are associated with an outcome-specific variable indicative of disease progression.

22. The system of claim 21, further comprising a component configured to suggest a treatment strategy based on the outcome-specific variable.

23. The system of claim 21, wherein the tissue sample is a tumor sample, and wherein the outcome-specific variable is indicative of metastatic potential.

24. The system of claim 21, further comprising a component configured to guide treatment strategies based on identifying cell types and activation and evolutionary states based on the weighted graph.

25. The system of claim 13, wherein the plurality of multicellular in vitro models is selected from 2D cultures, 3D spheroids, organoids, and biomimetic microphysiological systems.

26. A method for generating a spatially indicative representation of heterogeneous cellular communication from multi-parameter cellular and subcellular imaging data obtained from multiple tissue samples from multiple patients or multiple multicellular in vitro models, the method comprising: generating a quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data, and identifying microregions of one of the tissue samples based on the quantification, including phenotyping the multi-parameter cellular and subcellular imaging data to identify certain different predetermined phenotypes; and constructing a communication graph for a microregion, wherein each phenotype is a node in the communication graph, and wherein an edge between each pair of phenotypes in the communication graph indicates the effect of one phenotype in the pair on the presence of the other phenotype in the pair, wherein each phenotype in the communication graph is represented by a data vector obtained from the multi-parameter cellular and subcellular imaging data, and wherein the method comprises: For each pair of phenotypes, a numerical relationship is established between the pair of phenotypes by determining the strength of association or influence value between the data vectors of the pair of phenotypes, and a directionality is established for each numerical relationship, wherein the edge between each pair of phenotypes indicates the numerical relationship between the pair of phenotypes and the directionality of the numerical relationship between the pair of phenotypes.

27. The method of claim 26, wherein each association strength or influence value is a linear or non-linear correlation coefficient value.

28. The method of claim 26, wherein each data vector is a vector of expression values ​​for a plurality of predetermined biomarkers, wherein the predetermined biomarkers are used to generate the multi-parameter cellular and subcellular imaging data.

29. The method of claim 28, wherein each vector of expression values ​​is a biomarker intensity pattern for a plurality of predetermined biomarkers.

30. The method of claim 28, wherein establishing the directionality for each of the numerical relationships is based on a plurality of receptor-ligand databases for the predetermined biomarkers.

31. The method of claim 26, further comprising interpreting each data vector to determine a plurality of activation states for each phenotype, and including each of the plurality of activation states in the communication graph.

32. The method of claim 26, wherein determining the strength of association or effect size between the data vectors for the pair of phenotypes comprises removing confounding effects of phenotypes other than the phenotypes in the pair.

33. The method of claim 26, wherein the plurality of multicellular in vitro models is selected from the group consisting of 2D cultures, 3D spheroids, organoids, and biomimetic microphysiological systems.

34. A non-transitory computer readable medium storing one or more programs including instructions which, when executed by a computer, cause the computer to perform the method according to claim 26.

35. A computerized system for generating a spatially indicative heterogeneous cellular communication representation from multi-parameter cellular and subcellular imaging data obtained from a plurality of tissue samples from a plurality of patients or a plurality of multi-cellular in vitro models, the computerized system comprising: A processing device, wherein the processing device comprises: a spatial heterogeneity quantification component configured to generate a quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data, including phenotyping the multi-parameter cellular and subcellular imaging data to identify certain different predetermined phenotypes; a microregion identification component configured to identify a plurality of microregions of the plurality of tissue samples based on the quantification, each microregion being associated with one of the tissue samples; and a communication network component configured to construct a communication graph for one of the selected microregions, wherein each phenotype is a node in the communication graph, and wherein an edge between each pair of phenotypes in the communication graph indicates an effect of one phenotype in the pair on the presence of another phenotype in the pair, wherein each phenotype in the communication graph is represented by a data vector obtained from the multi-parameter cellular and subcellular imaging data, and wherein the communication network component is configured to: For each pair of phenotypes, a numerical relationship is established between the pair of phenotypes by determining the strength of association or influence value between the data vectors of the pair of phenotypes, and a directionality is established for each numerical relationship, wherein the edge between each pair of phenotypes indicates the numerical relationship between the pair of phenotypes and the directionality of the numerical relationship between the pair of phenotypes.

36. The system of claim 35, wherein each association strength or influence value is a linear or non-linear correlation coefficient value.

37. The system of claim 35, wherein each data vector is a vector of expression values ​​for a plurality of predetermined biomarkers, wherein the predetermined biomarkers are used to generate the multi-parameter cellular and subcellular imaging data.

38. The system of claim 37, wherein each expression value vector is a biomarker intensity pattern for a plurality of predetermined biomarkers.

39. The system of claim 37, wherein establishing directionality for each numerical relationship is based on a plurality of receptor-ligand databases for the predetermined biomarker.

40. The system of claim 35, wherein the communication network component is further configured to interpret each data vector to determine a plurality of activation states for each phenotype and include each of the plurality of activation states in the communication graph.

41. The system of claim 35, wherein determining the strength of association or effect size between the data vectors for the pair of phenotypes comprises removing confounding effects of phenotypes other than the phenotypes in the pair.

42. The system of claim 35, wherein the plurality of multicellular in vitro models is selected from 2D cultures, 3D spheroids, organoids, and biomimetic microphysiological systems.

43. A method of representing the temporal evolution of disease progression in a specific patient, wherein the specific patient is one of a plurality of patients, and wherein the plurality of patients are associated with multi-parameter cellular and subcellular imaging data of a plurality of tissue samples obtained from the plurality of patients, the method comprising: A geometric representation of a disease landscape is generated for a specific patient, wherein the geometric representation includes a plurality of points, wherein each point on the geometric representation: (i) describes the disease state of the specific patient at a specific time and is based on a specific one of the tissue samples associated with the selected patient, (ii) is based on a microregion in the specific one of the tissue samples based on multi-parameter cellular and subcellular imaging data, wherein each microregion includes a plurality of phenotypes, and (iii) includes a heterogeneous cellular communication graph for the microregion, which includes a spatially indicated heterogeneous cellular communication representation for the microregion, wherein each phenotype of the microregion is a node in the communication graph, and wherein an edge between each pair of phenotypes in the communication graph indicates an influence of one phenotype in the pair on the presence of another phenotype in the pair, wherein each phenotype in the communication graph is represented by a data vector obtained from the multi-parameter cellular and subcellular imaging data, and wherein the method includes: for each pair of phenotypes, establishing a numerical relationship between the pair of phenotypes by determining an association strength or influence value between the data vectors of the pair of phenotypes, and establishing a directionality for each numerical relationship, wherein each edge between the pair of phenotypes indicates the numerical relationship between the pair of phenotypes and the directionality of the numerical relationship between the pair of phenotypes.

44. The method of claim 43, wherein each heterogeneous cellular communication graph is associated with a systems biology model comprising a system of ordinary differential equations.

45. The method of claim 44, wherein for each heterogeneous cellular communication graph, the system of ordinary differential equations defines the dynamics of the disease landscape to predict the temporal evolution of the disease landscape for the specific patient.

46. ​​A method according to claim 43, wherein each microregion is identified by generating a quantification of spatial heterogeneity between cells of certain different predetermined phenotypes in the multi-parameter cellular and subcellular imaging data and identifying the microregion based on the quantification, comprising cellular phenotyping the multi-parameter cellular and subcellular imaging data to identify certain different predetermined phenotypes.

Images