Integrating knowledge and omics to decipher mechanisms via large‐scale models of signaling networks

Abstract

Signal transduction governs cellular behavior, and its dysregulation often leads to human disease. To understand this process, we can use network models based on prior knowledge, where nodes represent biomolecules, usually proteins, and edges indicate interactions between them. Several computational methods combine untargeted omics data with prior knowledge to estimate the state of signaling networks in specific biological scenarios. Here, we review, compare, and classify recent network approaches according to their characteristics in terms of input omics data, prior knowledge and underlying methodologies. We highlight existing challenges in the field, such as the general lack of ground truth and the limitations of prior knowledge. We also point out new omics developments that may have a profound impact, such as single-cell proteomics or large-scale profiling of protein conformational changes. We provide both an introduction for interested users seeking strategies to study cell signaling on a large scale and an update for seasoned modelers.

Advancing Targeted Protein Degradation via Multiomics Profiling and Artificial Intelligence

“…In particular, functional ontologies have been mapped to microscopy images, 214 and mechanistic models of transcriptional readouts are becoming established tools. 215 In summary, we believe that the necessary computational elements are already in-place for maximized, indepth interpretation of phenotypic readouts. We anticipate that future degraders will frequently be discovered by phenotypic assays able to illuminate corners of the biological space that have not yet been explored with small-molecule perturbations.…”

“…While similarity-based interpretation of transcriptomics and imaging profiles is likely to remain the by-default analytical framework for phenotypic screens, recent AI/ML-based approaches point toward standalone or direct mechanistic interpretation of phenotypic readouts. In particular, functional ontologies have been mapped to microscopy images, and mechanistic models of transcriptional readouts are becoming established tools . In summary, we believe that the necessary computational elements are already in-place for maximized, in-depth interpretation of phenotypic readouts.…”

Application of perturbation gene expression profiles in drug discovery—From mechanism of action to quantitative modelling

“…Importantly, virtual perturbation in these models can help to identify biomarkers and synergistic drug combinations (Eduati et al, 2017). These models generally use some prior-knowledge biological (signalling) network (Türei et al, 2021) to connect proteins (nodes) and use data-driven methods to parameterise (fit) the network parameters (edge weights) to the biological context (Garrido-Rodriguez et al, 2022). To fit the network parameters, a wide range of computational tools are used, like graph algorithms (Browaeys et al, 2019), integer linear programming (Liu et al, 2019) or neural network architectures (Yuan et al, 2021;Nilsson et al, 2022).…”

Network Modeling and Control of Dynamic Disease Pathways, Review and Perspectives

“…A biological network is composed of bio-molecules and the interactions or reactions between the bio-molecules [6]. Biological networks are used to conceptually represent signaling, gene regulation and metabolism [7]. Biological networks are useful to disentangle biological mechanisms, to study etiology, and to predict therapeutic responses [8].…”