Multi-omics protein-coding units as massively parallel Bayesian networks: Empirical validation of causality structure

Alberto Zenere; Olof Rundquist; Mika Gustafsson; Claudio Altafini

doi:10.1016/j.isci.2022.104048

Multi-omics protein-coding units as massively parallel Bayesian networks: Empirical validation of causality structure

iScience. 2022 Mar 11;25(4):104048. doi: 10.1016/j.isci.2022.104048. eCollection 2022 Apr 15.

Authors

Alberto Zenere¹, Olof Rundquist², Mika Gustafsson², Claudio Altafini¹

Affiliations

¹ Division of Automatic Control, Department of Electrical Engineering, Linköping University, SE-58183 Linköping, Sweden.
² Bioinformatics, Department of Physics, Chemistry and Biology, Linköping University, SE-58183 Linköping, Sweden.

Abstract

In this article we use high-throughput epigenomics, transcriptomics, and proteomics data to construct fine-graded models of the "protein-coding units" gathering all transcript isoforms and chromatin accessibility peaks associated with more than 4000 genes in humans. Each protein-coding unit has the structure of a directed acyclic graph (DAG) and can be represented as a Bayesian network. The factorization of the joint probability distribution induced by the DAGs imposes a number of conditional independence relationships among the variables forming a protein-coding unit, corresponding to the missing edges in the DAGs. We show that a large fraction of these conditional independencies are indeed verified by the data. Factors driving this verification appear to be the structural and functional annotation of the transcript isoforms, as well as a notion of structural balance (or frustration-free) of the corresponding sample correlation graph, which naturally leads to reduction of correlation (and hence to independence) upon conditioning.

Keywords: Bioinformatics; Mathematical biosciences; Omics; Systems biology.