Background

Cell signalling describes the processes that occur in the cell in response to changes in its environment. They are based on complex protein interaction networks, the wiring of which can differ depending on the cell state, type, its environment and many other parameters, leading also to distinct phenotypes. Deregulation of cell signalling is the basis of many human diseases including cancer. Therefore, understanding human cell signalling both at the molecular and the systems level is critical for understanding how our cells work but also for tackling disease.

For this project, we bring together expertise in human cell signaling, and large-scale, integrative computational modeling. Our aim is to use techniques derived from “whole-cell” models of bacteria to produce a comprehensive model of cell signaling in humans. Whole cell computational models provide a holistic view of cell function. They are able to consider the context in which each cell process occurs and allow prediction of the effect of specific perturbations on cell phenotype. They therefore provide the ideal strategy to study human context-specific signalling and how it regulates distinct phenotypes.

Project

A large-scale, integrative model of human cell signalling

In this project we aim to create a predictive model of human cell signalling using data-driven cell signalling modules and hybrid approaches that combine network propagation with executable modelling.

A major challenge in creating a whole cell human cell signalling model, is our very limited detailed knowledge of the system – with the exception of a very small fraction of the human cell signalling network, that has been extensively studied. To mitigate this, the Petsalaki group uses public multi-omics datasets and machine learning to identify data-driven modules of human cell signalling (unpublished). These will act as units of cell signalling and the basis of the whole cell model. The Covert group has developed multiple tools that allow linking of heterogeneous modules in a functional way to generate large executable models. Most notably, through their efforts to build “whole-cell” models of bacteria – most recently in E. coli – they have developed a software framework, Vivarium, a software tool for building integrative multiscale models.

Vivarium provides an interface that can make any mechanistic model into a module that can be wired together into larger composite models, and then parallelized and run across multiple CPUs with Vivarium’s simulation engine. This project will build on existing and ongoing work of the two collaborating groups, to generate a large-scale, integrative model of human cell signalling.  This effort could include optimising module identification; developing approaches that allow signal propagation within and across these modules using methods; including but not limited to network diffusion, boolean, logic or ODE modelling; and evaluating the model through in silico experiments that will then be validated with existing public datasets and/or in the lab. We expect this first-of-its-kind model to provide a new way of generating such integrative models in human cells, and also to provide insight into the regulation of human cell signalling, including the generation of multiple new hypotheses, predictions of context-specific cell responses and potential applications to synthetic biology. 

Publications:

<sup>1. Brandon M Invergo, Borgthor Petursson, Nosheen Akhtar, David Bradley, Girolamo Giudice, Maruan Hijazi, Pedro Cutillas, Evangelia Petsalaki, Pedro Beltrao, “Prediction of Signed Protein Kinase Regulatory Circuits”. Cell systems. 2020
2. Macklin DN, Ahn-Horst TA, Choi H, Ruggero NA, Carrera J, Mason JC, Sun G, Agmon E, DeFelice MM, Maayan I, Lane K, Spangler RK, Gillies TE, Paull ML, Akhter S, Bray SR, Weaver DS, Keseler IM, Karp PD, Morrison JH, Covert MW. Simultaneous cross-evaluation of heterogeneous E. coli datasets via mechanistic simulation. Science. 2020 
3. Eran Agmon, Ryan K. Spangler, Christopher J. Skalnik, William Poole, Shayn M Peirce, Jerry H. Morrison, Markus W. Covert. Vivarium: an interface and engine for integrative multiscale modeling in computational biology. bioRxiv. 2021</sup>

Find out more:

Do you want to develop predictive models of whole-cell signalling? Get in touch, we would love to hear from you!