A scalable method for parameter-free simulation and validation of mechanistic cellular signal transduction network models

Abstract

The metabolic modelling community has established the gold standard for bottom-up systems biology with reconstruction, validation and simulation of mechanistic genome-scale models. In contrast, current established methods do not support genome-scale mechanistic models of signal transduction networks. These networks encode information through internal states, and dealing with these states leads to scalability issues both in model formulation and execution. While rule based modelling can be used for efficient model definition (through rules) and simulation (through agent based execution), these quantitative models require parametrisation. This introduces yet another layer of uncertainty, due to the sparsity of reliably measured parameters. Hence, parameter-free simulation and validation will be important to support large-scale reconstruction and analysis of signal transduction network models. Here, we present a scalable method for parameterfree simulation of mechanistic signal transduction network models. It is based on rxncon, the reaction-contingency language, which describes the signalling network in terms of elemental reactions and states. We develop two generic update rules for states and reactions, based on detail analysis of two minimal reaction motifs, that can be used to map an arbitrary rxncon network on fully defined bipartite Boolean model. Locally defined update rules are assembled into a functional model without system level optimisation, making the methods suitable for network validation. Furthermore, an underlying model defined solely in terms of molecular reactions and causalities can be used to explain and predict system level behaviour. Taken together, we present a method for parameter-free simulation of mechanistic signal transduction models. Through scalable model definition and simulation, and the independence of quantitative parameters, it opens up for simulation and validation of mechanistic genome-scale models of signal transduction networks.

Modelling heterogeneous intracellular networks at the cellular scale

“…The assumptions central to constraint-based methods (e.g. assumptions of steady state and optimality of selected biological goals such as growth) are less applicable to other cellular processes, including signalling and transcriptional regulation, so detailed dynamic models of these networks instead rely on alternative formalisms such as ODE or Boolean models [14][15][16][17][18].…”

Automated verification, assembly, and extension of GBM stem cell network model with knowledge from literature and data

“…Current approaches for quantifying signaling network accuracy include validating dynamic network outcomes against experimental data or verifying the individual interactions with experimental data or previous knowledge [8]. For static network models, validating dynamic behavior is not possible.…”

Diverse Cell Stimulation Kinetics Identify Predictive Signal Transduction Models

“…A key obstacle that prevents predictive modeling of cell signaling is the gross mismatch between the preponderance of biological complexity and the sparsity of quantitative experimental data ( Handly et al., 2016 ; Janes and Lauffenburger, 2013 ; Vanhaelen et al., 2017 ). As a consequence, most current models of signal transduction pathways suffer from lack of dynamic richness in the data resulting in either too simple ( Adler and Alon, 2018 ; Csete and Doyle, 2002 ; Muzzey et al., 2009 ) or too complex ( Groβ et al., 2019 ; Klipp et al., 2005 ; Romers et al., 2020 ) models with limited predictive power ( Handly et al., 2016 ; Janes and Lauffenburger, 2013 ). To address the disparity between biological complexity and lack of richness in experimental data, one paradigm has been to devise experiments with higher content (e.g., sequencing or multiplexed single-cell imaging) or higher throughput (e.g., flow cytometry or parallelized microfluidics) in hope that large amounts of data will eventually fill the gap between mechanistic and predictive understanding ( Efremova et al., 2020 ; Labib and Kelley, 2020 ).…”