Bayesian calibration of a stochastic,multiscale agent-based model for predicting in vitro tumor growth

Hybrid multiscale agent-based models (ABMs) are unique in their ability to simulate individual cell interactions and microenvironmental dynamics. Unfortunately, the high computational cost of modeling individual cells, the inherent stochasticity of cell dynamics, and numerous model parameters are fundamental limitations of applying such models to predict tumor dynamics. To overcome these challenges, we have developed a coarse-grained two-scale ABM (cgABM) with a reduced parameter space that allows for an accurate and efficient calibration using a set of time-resolved microscopy measurements of cancer cells grown with different initial conditions. The multiscale model consists of a reaction-diffusion type model capturing the spatio-temporal evolution of glucose and growth factors in the tumor microenvironment (at tissue scale), coupled with a lattice-free ABM to simulate individual cell dynamics (at cellular scale). The experimental data consists of BT474 human breast carcinoma cells initialized with different glucose concentrations and tumor cell confluences. The confluence of live and dead cells was measured every three hours over four days. Given this model, we perform a time-dependent global sensitivity analysis to identify the relative importance of the model parameters. The subsequent cgABM is calibrated within a Bayesian framework to the experimental data to estimate model parameters, which are then used to predict the temporal evolution of the living and dead cell populations. To this end, a moment-based Bayesian inference is proposed to account for the stochasticity of the cgABM while quantifying uncertainties due to limited temporal observational data. The cgABM reduces the computational time of ABM simulations by 93% to 97% while staying within a 3% difference in prediction compared to ABM. Additionally, the cgABM can reliably predict the temporal evolution of breast cancer cells observed by the microscopy data with an average error and standard deviation for live and dead cells being 7.61±2.01 and 5.78±1.13, respectively.     

A hybrid multi-compartment model of granuloma formation and T cell priming in tuberculosis

Tuberculosis is a worldwide health problem with 2 billion people infected with Mycobacterium tuberculosis (Mtb, the bacteria causing TB). The hallmark of infection is the emergence of organized structures of immune cells forming primarily in the lung in response to infection. Granulomas physically contain and immunologically restrain bacteria that cannot be cleared. We have developed several models that spatially characterize the dynamics of the host-mycobacterial interaction, and identified mechanisms that control granuloma formation and development. In particular, we published several agent-based models (ABMs) of granuloma formation in TB that include many subtypes of T cell populations, macrophages as well as key cytokine and chemokine effector molecules. These ABM studies emphasize the important role of T-cell related mechanisms in infection progression, such as magnitude and timing of T cell recruitment, and macrophage activation. In these models, the priming and recruitment of T cells from the lung draining lymph node (LN) was captured phenomenologically. In addition to these ABM studies, we have also developed several multi-organ models using ODEs to examine trafficking of cells between, for example, the lung and LN. While we can predict temporal dynamic behaviors, those models are not coupled to the spatial aspects of granuloma. To this end, we have developed a multi-organ model that is hybrid: an ABM for the lung compartment and a non-linear system of ODE representing the lymph node compartment. This hybrid multi-organ approach to study TB granuloma formation in the lung and immune priming in the LN allows us to dissect protective mechanisms that cannot be achieved using the single compartment or multi-compartment ODE system. The main finding of this work is that trafficking of important cells known as antigen presenting cells from the lung to the lymph node is a key control mechanism for protective immunity: the entire spectrum of infection outcomes can be regulated by key immune cell migration rates. Our hybrid multi-organ implementation suggests that effector CD4+ T cells can rescue the system from a persistent infection and lead to clearance once a granuloma is fully formed. This could be effective as an immunotherapy strategy for latently infected individuals.     

Mechanobiological model of arterial growth and remodeling

A coupled agent-based model (ABM) and finite element analysis (FEA) computational framework is developed to study the interplay of bio-chemo-mechanical factors in blood vessels and their role in maintaining homeostasis. The agent-based model implements the power of REPAST Simphony libraries and adapts its environment for biological simulations. Coupling a continuum-level model (FEA) to a cellular-level model (ABM) has enabled this computational framework to capture the response of blood vessels to increased or decreased levels of growth factors, proteases and other signaling molecules (on the micro scale) as well as altered blood pressure. Performance of the model is assessed by simulating porcine left anterior descending artery under normotensive conditions and transient increases in blood pressure and by analyzing sensitivity of the model to variations in the rule parameters of the ABM. These simulations proved that the model is stable under normotensive conditions and can recover from transient increases in blood pressure. Sensitivity studies revealed that the model is most sensitive to variations in the concentration of growth factors that affect cellular proliferation and regulate extracellular matrix composition (mainly collagen).     

An agent-based model of sleeping sickness: simulation trials of a forest focus in southern Cameroon

An agent-based model (AMB) used to simulate the spread of Human African Trypanosomiasis is presented together with the results of simulations of a focus of the disease. This model is a completely spatialized approach taking into account a series of often overlooked parameters such as human behaviour (activity-related movements), the density and mobility of the disease vectors--tsetse flies (Glossina spp.)--and the influence of other tsetse feeding hosts (livestock and wild animal populations). The agents that represent humans and tsetse flies move in a spatially structured environment managed by specialized location agents. Existing compartmental mathematical models governed by differential equations fail to incorporate the spatial dimension of the disease transmission. Furthermore, on a small scale, transmission is unrealistically represented by entities less than one. This ABM was tested with data from one village of the Bipindi sleeping sickness focus (southern Cameroon) and with obtained realistic simulations of stable transmission involving an animal reservoir. In varying different spatial configurations, we observe that the stability of spread is linked to the spatial complexity (number of heterogeneous locations). The prevalence is very sensitive to the human densities and to the number of tsetse flies initially infected in a given location. A relatively low and durable prevalence is obtained with shortening the phase I. In addition, we discuss some upgrading possibilities, in particular the linkage to a Geographical Information System (GIS). The agent-based approach offers new ways to understanding the spread of the disease and a tool to evaluate risk and test control strategies.     

An “Age” Structured Model of Hematopoietic Stem Cell Organization with Application to Chronic Myeloid Leukemia

Previously, we have modeled hematopoietic stem cell organization by a stochastic, single cell-based approach. Applications to different experimental systems demonstrated that this model consistently explains a broad variety of in vivo and in vitro data. A major advantage of the agent-based model (ABM) is the representation of heterogeneity within the hematopoietic stem cell population. However, this advantage comes at the price of time-consuming simulations if the systems become large. One example in this respect is the modeling of disease and treatment dynamics in patients with chronic myeloid leukemia (CML), where the realistic number of individual cells to be considered exceeds 10⁶. To overcome this deficiency, without losing the representation of the inherent heterogeneity of the stem cell population, we here propose to approximate the ABM by a system of partial differential equations (PDEs). The major benefit of such an approach is its independence from the size of the system. Although this mean field approach includes a number of simplifying assumptions compared to the ABM, it retains the key structure of the model including the “age”-structure of stem cells. We show that the PDE model qualitatively and quantitatively reproduces the results of the agent-based approach.     

A multi-approach and multi-scale platform to model CD4+ T cells responding to infections

Immune responses rely on a complex adaptive system in which the body and infections interact at multiple scales and in different compartments. We developed a modular model of CD4+ T cells, which uses four modeling approaches to integrate processes at three spatial scales in different tissues. In each cell, signal transduction and gene regulation are described by a logical model, metabolism by constraint-based models. Cell population dynamics are described by an agent-based model and systemic cytokine concentrations by ordinary differential equations. A Monte Carlo simulation algorithm allows information to flow efficiently between the four modules by separating the time scales. Such modularity improves computational performance and versatility and facilitates data integration. We validated our technology by reproducing known experimental results, including differentiation patterns of CD4+ T cells triggered by different combinations of cytokines, metabolic regulation by IL2 in these cells, and their response to influenza infection. In doing so, we added multi-scale insights to single-scale studies and demonstrated its predictive power by discovering switch-like and oscillatory behaviors of CD4+ T cells that arise from nonlinear dynamics interwoven across three scales. We identified the inflamed lymph node’s ability to retain naive CD4+ T cells as a key mechanism in generating these emergent behaviors. We envision our model and the generic framework encompassing it to serve as a tool for understanding cellular and molecular immunological problems through the lens of systems immunology.     

Novel Multiscale Modeling Tool Applied to Pseudomonas aeruginosa Biofilm Formation

Multiscale modeling is used to represent biological systems with increasing frequency and success. Multiscale models are often hybrids of different modeling frameworks and programming languages. We present the MATLAB-NetLogo extension (MatNet) as a novel tool for multiscale modeling. We demonstrate the utility of the tool with a multiscale model of Pseudomonas aeruginosa biofilm formation that incorporates both an agent-based model (ABM) and constraint-based metabolic modeling. The hybrid model correctly recapitulates oxygen-limited biofilm metabolic activity and predicts increased growth rate via anaerobic respiration with the addition of nitrate to the growth media. In addition, a genome-wide survey of metabolic mutants and biofilm formation exemplifies the powerful analyses that are enabled by this computational modeling tool.     

Parameter estimation and sensitivity analysis in an agent-based model of Leishmania major infection

Computer models of disease take a systems biology approach toward understanding host-pathogen interactions. In particular, data driven computer model calibration is the basis for inference of immunological and pathogen parameters, assessment of model validity, and comparison between alternative models of immune or pathogen behavior. In this paper we describe the calibration and analysis of an agent-based model of Leishmania major infection. A model of macrophage loss following uptake of necrotic tissue is proposed to explain macrophage depletion following peak infection. Using Gaussian processes to approximate the computer code, we perform a sensitivity analysis to identify important parameters and to characterize their influence on the simulated infection. The analysis indicates that increasing growth rate can favor or suppress pathogen loads, depending on the infection stage and the pathogen's ability to avoid detection. Subsequent calibration of the model against previously published biological observations suggests that L. major has a relatively slow growth rate and can replicate for an extended period of time before damaging the host cell.     

Modeling Protective Anti-Tumor Immunity via Preventative Cancer Vaccines Using a Hybrid Agent-based and Delay Differential Equation Approach

A next generation approach to cancer envisions developing preventative vaccinations to stimulate a person''s immune cells, particularly cytotoxic T lymphocytes (CTLs), to eliminate incipient tumors before clinical detection. The purpose of our study is to quantitatively assess whether such an approach would be feasible, and if so, how many anti-cancer CTLs would have to be primed against tumor antigen to provide significant protection. To understand the relevant dynamics, we develop a two-compartment model of tumor-immune interactions at the tumor site and the draining lymph node. We model interactions at the tumor site using an agent-based model (ABM) and dynamics in the lymph node using a system of delay differential equations (DDEs). We combine the models into a hybrid ABM-DDE system and investigate dynamics over a wide range of parameters, including cell proliferation rates, tumor antigenicity, CTL recruitment times, and initial memory CTL populations. Our results indicate that an anti-cancer memory CTL pool of 3% or less can successfully eradicate a tumor population over a wide range of model parameters, implying that a vaccination approach is feasible. In addition, sensitivity analysis of our model reveals conditions that will result in rapid tumor destruction, oscillation, and polynomial rather than exponential decline in the tumor population due to tumor geometry.     

Modeling mechanical inhomogeneities in small populations of proliferating monolayers and spheroids

Understanding the mechanical behavior of multicellular monolayers and spheroids is fundamental to tissue culture, organism development, and the early stages of tumor growth. Proliferating cells in monolayers and spheroids experience mechanical forces as they grow and divide and local inhomogeneities in the mechanical microenvironment can cause individual cells within the multicellular system to grow and divide at different rates. This differential growth, combined with cell division and reorganization, leads to residual stress. Multiple different modeling approaches have been taken to understand and predict the residual stresses that arise in growing multicellular systems, particularly tumor spheroids. Here, we show that by using a mechanically robust agent-based model constructed with the peridynamic framework, we gain a better understanding of residual stresses in multicellular systems as they grow from a single cell. In particular, we focus on small populations of cells (1–100 s) where population behavior is highly stochastic and prior investigation has been limited. We compare the average strain energy density of cells in monolayers and spheroids using different growth and division rules and find that, on average, cells in spheroids have a higher strain energy density than cells in monolayers. We also find that cells in the interior of a growing spheroid are, on average, in compression. Finally, we demonstrate the importance of accounting for stochastic fluctuations in the mechanical environment, particularly when the cellular response to mechanical cues is nonlinear. The results presented here serve as a starting point for both further investigation with agent-based models, and for the incorporation of major findings from agent-based models into continuum scale models when explicit representation of individual cells is not computationally feasible.     

From actors to agents in socio-ecological systems models

The ecosystem service concept has emphasized the role of people within socio-ecological systems (SESs). In this paper, we review and discuss alternative ways of representing people, their behaviour and decision-making processes in SES models using an agent-based modelling (ABM) approach. We also explore how ABM can be empirically grounded using information from social survey. The capacity for ABM to be generalized beyond case studies represents a crucial next step in modelling SESs, although this comes with considerable intellectual challenges. We propose the notion of human functional types, as an analogy of plant functional types, to support the expansion (scaling) of ABM to larger areas. The expansion of scope also implies the need to represent institutional agents in SES models in order to account for alternative governance structures and policy feedbacks. Further development in the coupling of human-environment systems would contribute considerably to better application and use of the ecosystem service concept.     

Pathology of spontaneous and immunotherapy‐induced tumor regression in a murine model of melanoma

We evaluated the spontaneous and immunotherapy‐induced histological changes in the tumor microenvironment of a mouse melanoma regression model consisting of immunocompetent C57BL/6J mice implanted with syngeneic YUMMER1.7 melanoma cells. We focused on tumor regression phenotypes and spatial relationships of melanoma cells with B cells and neutrophils since this was not previously described. We found common themes to the host response to cancer irrespective of the mode of tumor regression. In nonregression tumors, melanoma cells were epithelioid shaped and tightly packed. In regression tumors, melanoma cells were spindle shaped and discohesive. B cells including plasmablasts and plasma cells were numerous and were increased with immunotherapy. Neutrophils were in direct contact with dead or dying melanoma cells. Immunotherapy increased neutrophil counts and induced neutrophil extracellular traps (NETs)‐like formations and geographic necrosis. Beyond tumor regression, the increase in the B cell and neutrophil response could play a role in immunotherapy‐induced adverse reactions.     

Spatial scaling in multiscale models: methods for coupling agent-based and finite-element models of wound healing

Biomechanics and Modeling in Mechanobiology - Multiscale models that couple agent-based modeling (ABM) and finite-element modeling (FEM) allow the dynamic simulation of tissue remodeling and wound...     

An agent-based model of inflammation and fibrosis following particulate exposure in the lung

Inflammation and airway remodeling occur in a variety of airway diseases. Modeling aspects of the inflammatory and fibrotic processes following repeated exposure to particulate matter may provide insights into a spectrum of airway diseases, as well as prevention/treatment strategies. An agent-based model (ABM) was created to examine the response of an abstracted population of inflammatory cells (nominally macrophages, but possibly including other inflammatory cells such as lymphocytes) and cells involved in remodeling (nominally fibroblasts) to particulate exposure. The model focused on a limited number of relevant interactions, specifically those among macrophages, fibroblasts, a pro-inflammatory cytokine (TNF-α), an anti-inflammatory cytokine (TGF-β1), collagen deposition, and tissue damage. The model yielded three distinct states that were equated with (1) self-resolving inflammation and a return to baseline, (2) a pro-inflammatory process of localized tissue damage and fibrosis, and (3) elevated pro- and anti-inflammatory cytokines, persistent tissue damage, and fibrosis outcomes. Experimental results consistent with these predicted states were observed in histology sections of lung tissue from mice exposed to particulate matter. Systematic in silico studies suggested that the development of each state depended primarily upon the degree and duration of exposure. Thus, a relatively simple ABM resulted in several, biologically feasible, emergent states, suggesting that the model captures certain salient features of inflammation following exposure of the lung to particulate matter. This ABM may hold future utility in the setting of airway disease resulting from inflammation and fibrosis following particulate exposure.     

Agent-based modelling as scientific method: a case study analysing primate social behaviour

A scientific methodology in general should provide two things: first, a means of explanation and, second, a mechanism for improving that explanation. Agent-based modelling (ABM) is a method that facilitates exploring the collective effects of individual action selection. The explanatory force of the model is the extent to which an observed meta-level phenomenon can be accounted for by the behaviour of its micro-level actors. This article demonstrates that this methodology can be applied to the biological sciences; agent-based models, like any other scientific hypotheses, can be tested, critiqued, generalized or specified. We review the state of the art for ABM as a methodology for biology and then present a case study based on the most widely published agent-based model in the biological sciences: Hemelrijk's DomWorld, a model of primate social behaviour. Our analysis shows some significant discrepancies between this model and the behaviour of the macaques, the genus used for our analysis. We also demonstrate that the model is not fragile: its other results are still valid and can be extended to compensate for these problems. This robustness is a standard advantage of experiment-based artificial intelligence modelling techniques over analytic modelling.     

Evolutionary game theory in an agent-based brain tumor model: exploring the 'Genotype-Phenotype' link

To investigate the genotype-phenotype link in a polyclonal cancer cell population, here we introduce evolutionary game theory into our previously developed agent-based brain tumor model. We model the heterogeneous cell population as a mixture of two distinct genotypes: the more proliferative Type A and the more migratory Type B. Our agent-based simulations reveal a phase transition in the tumor's velocity of spatial expansion linking the tumor fitness to genotypic composition. Specifically, velocity initially falls as rising payoffs reward the interactions among the more stationary Type A cells, but unexpectedly accelerates again when these A-A payoffs increase even further. At this latter accelerating stage, fewer migratory Type B cells appear to confer a competitive advantage in terms of the tumor's spatial aggression over the overall numerically dominating Type A cells, which in turn leads to an acceleration of the overall tumor dynamics while its surface roughness declines. We discuss potential implications of our findings for cancer research.     

Combined serial section-based 3D reconstruction of cervical carcinoma invasion using H&E/p16INK4a/CD3 alternate staining.

BACKGROUND: Malignant growth and invasiveness of cancers is a function of both intratumoral and stromal factors. The accessibility to nutrients, oxygen and growth factors, the stromal composition, and the interference with the immune system all shape the tumor invasion front. A recent study has shown a prognostic difference with respect to different invasion patterns analyzed on histological specimens of cervical cancers. The present study analyzes the spatial organization of a cervical cancer and the relation of the tumor invasion front and the infiltration with CD3(+) T-cells. METHODS: From a cervical squamous cell carcinoma specimen, 84 serial sections were performed and three interleaving series were stained with hematoxylin/eosin and immunohistochemistry directed against the cervical carcinoma biomarker p16(INK4a) and the T-cell marker CD3. Sections were passed through an image processing chain to obtain a reconstructed and segmented tissue volume. For local tumor invasion front analysis the mean curvature was used, which in turn was related to the respective local minimum tumor to T-cell distance as well to a T-cell originated diffusing substance's concentration at the tumor surface. RESULTS: Spatial models of the tumor tissue and the infiltrating T-cells were computed. The overall discrete compactness of the tumor invasion front was 0.89, corresponding to a pathological assessment of diffuse infiltration. The comparison of the tumor invasion front with the density of T-cell infiltration revealed an increased smoothening in regions with high T-cell infiltration. CONCLUSIONS: We could demonstrate the spatial organization of a cervical cancer and model the interaction between infiltrating T-cells with the tumor invasion front shape. Increased smoothening in regions with high T-cell infiltration suggests that T-cells may have an influence on the shaping of the tumor invasion front, e.g., by attacking tumor cells displaying specific antigens. The applied technique allows visualization of the spatial organization of tissues and could be extended to analyze multiple stains on alternating sections.     

Development of an Agent-Based Model (ABM) to Simulate the Immune System and Integration of a Regression Method to Estimate the Key ABM Parameters by Fitting the Experimental Data

Agent-based models (ABM) and differential equations (DE) are two commonly used methods for immune system simulation. However, it is difficult for ABM to estimate key parameters of the model by incorporating experimental data, whereas the differential equation model is incapable of describing the complicated immune system in detail. To overcome these problems, we developed an integrated ABM regression model (IABMR). It can combine the advantages of ABM and DE by employing ABM to mimic the multi-scale immune system with various phenotypes and types of cells as well as using the input and output of ABM to build up the Loess regression for key parameter estimation. Next, we employed the greedy algorithm to estimate the key parameters of the ABM with respect to the same experimental data set and used ABM to describe a 3D immune system similar to previous studies that employed the DE model. These results indicate that IABMR not only has the potential to simulate the immune system at various scales, phenotypes and cell types, but can also accurately infer the key parameters like DE model. Therefore, this study innovatively developed a complex system development mechanism that could simulate the complicated immune system in detail like ABM and validate the reliability and efficiency of model like DE by fitting the experimental data.     

Capturing the Dynamics of a Hybrid Multiscale Cancer Model with a Continuum Model

Cancer is a complex disease involving processes at spatial scales from subcellular, like cell signalling, to tissue scale, such as vascular network formation. A number of multiscale models have been developed to study the dynamics that emerge from the coupling between the intracellular, cellular and tissue scales. Here, we develop a continuum partial differential equation model to capture the dynamics of a particular multiscale model (a hybrid cellular automaton with discrete cells, diffusible factors and an explicit vascular network). The purpose is to test under which circumstances such a continuum model gives equivalent predictions to the original multiscale model, in the knowledge that the system details are known, and differences in model results can be explained in terms of model features (rather than unknown experimental confounding factors). The continuum model qualitatively replicates the dynamics from the multiscale model, with certain discrepancies observed owing to the differences in the modelling of certain processes. The continuum model admits travelling wave solutions for normal tissue growth and tumour invasion, with similar behaviour observed in the multiscale model. However, the continuum model enables us to analyse the spatially homogeneous steady states of the system, and hence to analyse these waves in more detail. We show that the tumour microenvironmental effects from the multiscale model mean that tumour invasion exhibits a so-called pushed wave when the carrying capacity for tumour cell proliferation is less than the total cell density at the tumour wave front. These pushed waves of tumour invasion propagate by triggering apoptosis of normal cells at the wave front. Otherwise, numerical evidence suggests that the wave speed can be predicted from linear analysis about the normal tissue steady state.     

A network-patch methodology for adapting agent-based models for directly transmitted disease to mosquito-borne disease

Mosquito-borne diseases cause significant public health burden and are widely re-emerging or emerging. Understanding, predicting, and mitigating the spread of mosquito-borne disease in diverse populations and geographies are ongoing modelling challenges. We propose a hybrid network-patch model for the spread of mosquito-borne pathogens that accounts for individual movement through mosquito habitats, extending the capabilities of existing agent-based models (ABMs) to include vector-borne diseases. The ABM are coupled with differential equations representing ‘clouds’ of mosquitoes in patches accounting for mosquito ecology. We adapted an ABM for humans using this method and investigated the importance of heterogeneity in pathogen spread, motivating the utility of models of individual behaviour. We observed that the final epidemic size is greater in patch models with a high risk patch frequently visited than in a homogeneous model. Our hybrid model quantifies the importance of the heterogeneity in the spread of mosquito-borne pathogens, guiding mitigation strategies.