Multi-scale computational modeling of developmental biology

An integrative multiscale view of early cardiac looping

The heart is the first organ to form and function during the development of an embryo. Heart development consists of a series of events believed to be highly conserved in vertebrates. Development of heart begins with the formation of the cardiac fields followed by a linear heart tube formation. The straight heart tube then undergoes a ventral bending prior to further bending and helical torsion to form a looped heart. The looping phase is then followed by ballooning, septation, and valve formation giving rise to a four-chambered heart in avians and mammals. The looping phase plays a central role in heart development. Successful looping is essential for proper alignment of the future cardiac chambers and tracts. As aberrant looping results in various congenital heart diseases, the mechanisms of cardiac looping have been studied for several decades by various disciplines. Many groups have studied anatomy, biology, genetics, and mechanical processes during heart looping, and have proposed multiple mechanisms. Computational modeling approaches have been utilized to examine the proposed mechanisms of the looping process. Still, the exact underlying mechanism(s) controlling the looping phase remain poorly understood. Although further experimental measurements are obviously still required, the need for more integrative computational modeling approaches is also apparent in order to make sense of the vast amount of experimental data and the complexity of multiscale developmental systems. Indeed, there needs to be an iterative interaction between experimentation and modeling in order to properly find the gap in the existing data and to validate proposed hypotheses. This article is categorized under: Cardiovascular Diseases > Genetics/Genomics/Epigenetics Cardiovascular Diseases > Computational Models Cardiovascular Diseases > Molecular and Cellular Physiology.

Predicting pattern formation in embryonic stem cells using a minimalist, agent-based probabilistic model

The mechanisms of pattern formation during embryonic development remain poorly understood. Embryonic stem cells in culture self-organise to form spatial patterns of gene expression upon geometrical confinement indicating that patterning is an emergent phenomenon that results from the many interactions between the cells. Here, we applied an agent-based modelling approach in order to identify plausible biological rules acting at the meso-scale within stem cell collectives that may explain spontaneous patterning. We tested different models involving differential motile behaviours with or without biases due to neighbour interactions. We introduced a new metric, termed stem cell aggregate pattern distance (SCAPD) to probabilistically assess the fitness of our models with empirical data. The best of our models improves fitness by 70% and 77% over the random models for a discoidal or an ellipsoidal stem cell confinement respectively. Collectively, our findings show that a parsimonious mechanism that involves differential motility is sufficient to explain the spontaneous patterning of the cells upon confinement. Our work also defines a region of the parameter space that is compatible with patterning. We hope that our approach will be applicable to many biological systems and will contribute towards facilitating progress by reducing the need for extensive and costly experiments.

Deep reinforcement learning of cell movement in the early stage of C.elegans embryogenesis

Motivation

Cell movement in the early phase of Caenorhabditis elegans development is regulated by a highly complex process in which a set of rules and connections are formulated at distinct scales. Previous efforts have demonstrated that agent-based, multi-scale modeling systems can integrate physical and biological rules and provide new avenues to study developmental systems. However, the application of these systems to model cell movement is still challenging and requires a comprehensive understanding of regulatory networks at the right scales. Recent developments in deep learning and reinforcement learning provide an unprecedented opportunity to explore cell movement using 3D time-lapse microscopy images.

Results

We present a deep reinforcement learning approach within an agent-based modeling system to characterize cell movement in the embryonic development of C.elegans. Our modeling system captures the complexity of cell movement patterns in the embryo and overcomes the local optimization problem encountered by traditional rule-based, agent-based modeling that uses greedy algorithms. We tested our model with two real developmental processes: the anterior movement of the Cpaaa cell via intercalation and the rearrangement of the superficial left-right asymmetry. In the first case, the model results suggested that Cpaaa's intercalation is an active directional cell movement caused by the continuous effects from a longer distance (farther than the length of two adjacent cells), as opposed to a passive movement caused by neighbor cell movements. In the second case, a leader-follower mechanism well explained the collective cell movement pattern in the asymmetry rearrangement. These results showed that our approach to introduce deep reinforcement learning into agent-based modeling can test regulatory mechanisms by exploring cell migration paths in a reverse engineering perspective. This model opens new doors to explore the large datasets generated by live imaging.

Availability and implementation

Source code is available at https://github.com/zwang84/drl4cellmovement.

Supplementary information

Supplementary data are available at Bioinformatics online.

eBrain: a Three Dimensional Simulation Tool to Study Drug Delivery in the Brain

Neurodegenerative disorders such as Alzheimer’s and Parkinson’s disease are severe disorders with acute symptoms that gradually progress. In the course of developing disease-modifying treatments for neurodegenerative disorders there is a need to develop novel strategies to increase efficacy of drugs and accelerate the development process. We developed a tool for simulating drug delivery in the brain by translating MRI data into an interactive 3D model. This tool, the eBrain, superimposes simulated drug diffusion and tissue uptake by inferring from the MRI data with a seamless display from any angle, magnification, or position. We discuss a representative implementation of eBrain that is inspired by clinical data in which insulin is intranasally administered to Alzheimer patients. Using extensive analysis of multiple eBrain simulations with varying parameters, we show the potential for eBrain to determine the optimal dosage to ensure drug delivery without overdosing the tissue. Specifically, we examined the efficacy of combined drug doses and potential compounds for tissue stimulation. Interestingly, our analysis uncovered that the drug efficacy is inferred from tissue intensity levels. Finally, we discuss the potential of eBrain and possible applications of eBrain to aid both inexperienced and experienced medical professionals as well as patients.

Perspective: The promise of multi-cellular engineered living systems

Recent technological breakthroughs in our ability to derive and differentiate induced pluripotent stem cells, organoid biology, organ-on-chip assays, and 3-D bioprinting have all contributed to a heightened interest in the design, assembly, and manufacture of living systems with a broad range of potential uses. This white paper summarizes the state of the emerging field of "multi-cellular engineered living systems," which are composed of interacting cell populations. Recent accomplishments are described, focusing on current and potential applications, as well as barriers to future advances, and the outlook for longer term benefits and potential ethical issues that need to be considered.

Agent-Based Modelling in Multicellular Systems Biology

This chapter aims at discussing the content of multi-agent based simulation (MABS) applied to computational biology i.e., to modelling and simulating biological systems by means of computational models, methodologies, and frameworks. In particular, the adoption of agent-based modelling (ABM) in the field of multicellular systems biology is explored, focussing on the challenging scenarios of developmental biology. After motivating why agent-based abstractions are critical in representing multicellular systems behaviour, MABS is discussed as the source of the most natural and appropriate mechanism for analysing the self-organising behaviour of systems of cells. As a case study, an application of MABS to the development of Drosophila Melanogaster is finally presented, which exploits the ALCHEMIST platform for agent-based simulation.

An Observation-Driven Agent-Based Modeling and Analysis Framework for C. elegans Embryogenesis

With cutting-edge live microscopy and image analysis, biologists can now systematically track individual cells in complex tissues and quantify cellular behavior over extended time windows. Computational approaches that utilize the systematic and quantitative data are needed to understand how cells interact in vivo to give rise to the different cell types and 3D morphology of tissues. An agent-based, minimum descriptive modeling and analysis framework is presented in this paper to study C. elegans embryogenesis. The framework is designed to incorporate the large amounts of experimental observations on cellular behavior and reserve data structures/interfaces that allow regulatory mechanisms to be added as more insights are gained. Observed cellular behaviors are organized into lineage identity, timing and direction of cell division, and path of cell movement. The framework also includes global parameters such as the eggshell and a clock. Division and movement behaviors are driven by statistical models of the observations. Data structures/interfaces are reserved for gene list, cell-cell interaction, cell fate and landscape, and other global parameters until the descriptive model is replaced by a regulatory mechanism. This approach provides a framework to handle the ongoing experiments of single-cell analysis of complex tissues where mechanistic insights lag data collection and need to be validated on complex observations.

Stochastic multi-scale models of competition within heterogeneous cellular populations: Simulation methods and mean-field analysis

We propose a modelling framework to analyse the stochastic behaviour of heterogeneous, multi-scale cellular populations. We illustrate our methodology with a particular example in which we study a population with an oxygen-regulated proliferation rate. Our formulation is based on an age-dependent stochastic process. Cells within the population are characterised by their age (i.e. time elapsed since they were born). The age-dependent (oxygen-regulated) birth rate is given by a stochastic model of oxygen-dependent cell cycle progression. Once the birth rate is determined, we formulate an age-dependent birth-and-death process, which dictates the time evolution of the cell population. The population is under a feedback loop which controls its steady state size (carrying capacity): cells consume oxygen which in turn fuels cell proliferation. We show that our stochastic model of cell cycle progression allows for heterogeneity within the cell population induced by stochastic effects. Such heterogeneous behaviour is reflected in variations in the proliferation rate. Within this set-up, we have established three main results. First, we have shown that the age to the G1/S transition, which essentially determines the birth rate, exhibits a remarkably simple scaling behaviour. Besides the fact that this simple behaviour emerges from a rather complex model, this allows for a huge simplification of our numerical methodology. A further result is the observation that heterogeneous populations undergo an internal process of quasi-neutral competition. Finally, we investigated the effects of cell-cycle-phase dependent therapies (such as radiation therapy) on heterogeneous populations. In particular, we have studied the case in which the population contains a quiescent sub-population. Our mean-field analysis and numerical simulations confirm that, if the survival fraction of the therapy is too high, rescue of the quiescent population occurs. This gives rise to emergence of resistance to therapy since the rescued population is less sensitive to therapy.

In-silico models of stem cell and developmental systems

Understanding how developmental systems evolve over time is a key question in stem cell and developmental biology research. However, due to hurdles of existing experimental techniques, our understanding of these systems as a whole remains partial and coarse. In recent years, we have been constructing in-silico models that synthesize experimental knowledge using software engineering tools. Our approach integrates known isolated mechanisms with simplified assumptions where the knowledge is limited. This has proven to be a powerful, yet underutilized, tool to analyze the developmental process. The models provide a means to study development in-silico by altering the model’s specifications, and thereby predict unforeseen phenomena to guide future experimental trials. To date, three organs from diverse evolutionary organisms have been modeled: the mouse pancreas, the C. elegans gonad, and partial rodent brain development. Analysis and execution of the models recapitulated the development of the organs, anticipated known experimental results and gave rise to novel testable predictions. Some of these results had already been validated experimentally. In this paper, I review our efforts in realistic in-silico modeling of stem cell research and developmental biology and discuss achievements and challenges. I envision that in the future, in-silico models as presented in this paper would become a common and useful technique for research in developmental biology and related research fields, particularly regenerative medicine, tissue engineering and cancer therapeutics.

Computational multiscale modeling of embryo development

Recent advances in live imaging and genetics of mammalian development which integrate observations of biochemical activity, cell-cell signaling and mechanical interactions between cells pave the way for predictive mathematical multi-scale modeling. In early mammalian embryo development, two of the most critical events which lead to tissue patterning involve changes in gene expression as well as mechanical interactions between cells. We discuss the relevance of mathematical modeling of multi-cellular systems and in particular in simulating these patterns and describe some of the technical challenges one encounters. Many of these issues are not unique for the embryonic system but are shared by other multi-cellular modeling areas.