Converting networks to predictive logic models from perturbation signalling data with CellNOpt

Abstract

Summary The molecular changes induced by perturbations such as drugs and ligands are highly informative of the intracellular wiring. Our capacity to generate large data-sets is increasing steadily. A useful way to extract mechanistic insight from the data is by integrating them with a prior knowledge network of signalling to obtain dynamic models. CellNOpt is a collection of Bioconductor R packages for building logic models from perturbation data and prior knowledge of signalling networks. We have recently developed new components and refined the existing ones in order to keep up with the computational demand of increasingly large data-sets, including (i) an efficient Integer Linear Programming (ILP), (ii) a probabilistic logic implementation for semi-quantitative datasets, (iii) the integration of a stochastic Boolean simulator, (iv) a tool to identify missing links, (v) systematic post-hoc analyses, and (vi) an R-Shiny tool to run CellNOpt interactively. Availability R-package(s): https://github.com/saezlab/cellnopt Supplementary information Supplementary data are available at Bioinformatics online.

In Silico Logical Modelling to Uncover Cooperative Interactions in Cancer

“…Meanwhile, the modelling formalism evolved to better account for the behaviour of regulatory networks [ 26 , 29 , 30 ]. Computational methods have been developed to map omics data to models, or to train logical models on data, thus enabling model contextualisation [ 31 , 32 , 33 , 34 ]. Here, we review representative examples of logical models reported in the literature that supported the identification of cooperative mechanisms in oncogenesis.…”

“…These models are somewhat “generic,” as they intend to represent an “average cell,” and thus do not account for tumours and patient heterogeneity. Hence, software tools, such as CellNOptR [ 34 , 46 ] and PRUNET [ 47 ], have been defined to contextualise models using high-throughput data, leading to cell type specific models (Ref. [ 47 ] provides an overview of these tools).…”

In silico logical modelling to uncover cooperative interactions in cancer

“…Meanwhile, the modelling formalism evolved to better account for the behaviour of regulatory networks [24,27,28]. Computational methods have been developed to map omics data to models, or to train logical models on data, thus enabling model contextualisation [29][30][31][32]. Here, we review representative examples of logical models reported in the literature that supported the identification of (which was not certified by peer review) is the author/funder.…”

“…These models are somewhat "generic" as they intend to represent an "average cell", and thus do not account for tumours and patient heterogeneity. Hence, software tools such as CellNOptR [32,43] and PRUNET [44], have been defined to contextualise models using high-throughput data, leading to cell type specific models ( [44] provides and overview of these tools). To confront logical models to clinical data, Béal et al recently proposed an approach integrating mutation data, copy number alterations, transcriptomic and proteomic data to models [31].…”

emba: R package for analysis and visualization of biomarkers in boolean model ensembles

“…The incorporated tools are accessed via a common programming interface (though originally implemented in different programming languages e.g. Java, Python, C++ and R) and offer a collection of features like accessing online model repositories (Helikar et al, 2012), model editing (Aurélien Naldi, Hernandez, Abou-Jaoudé, et al, 2018), dynamical analysis (finding attractors, stochastic simulations, reachability properties, model-checking techniques) (Klarner, Streck, Siebert, & Sahinalp, 2016;Müssel, Hopfensitz, & Kestler, 2010;Aurélien Naldi, 2018;Paulevé, 2017;Stoll et al, 2017) and model parameterization/optimization to fit perturbation signaling data (Gjerga et al, 2020;Terfve et al, 2012). Despite the diverse and multi-purpose logical modeling tools that exist, there is still a lack of data analysis-oriented software that assists with the discovery of predictive biomarkers in ensembles of parameterized boolean networks that have been subject to drug combination perturbations.…”