Abstract
Studying relationships among gene products by expression profile analysis is a common approach in systems biology. Many studies have generalized the outcomes to the different levels of central dogma information flow and assumed a correlation of transcript and protein expression levels. However, the relation between the various types of interaction (i.e., activation and inhibition) of gene products to their expression profiles has not been widely studied. In fact, looking for any perturbation according to differentially expressed genes is the common approach, while analyzing the effects of altered expression on the activity of signaling pathways is often ignored. In this study, we examine whether significant changes in gene expression necessarily lead to dysregulated signaling pathways. Using four commonly used and comprehensive databases, we extracted all relevant gene expression data and all relationships among directly linked gene pairs. We aimed to evaluate the ratio of coherency or sign consistency between the expression level as well as the causal relationships among the gene pairs. Through a comparison with random unconnected gene pairs, we illustrate that the signaling network is incoherent, and inconsistent with the recorded expression profile. Finally, we demonstrate that, to infer perturbed signaling pathways, we need to consider the type of relationships in addition to gene-product expression data, especially at the transcript level. We assert that identifying enriched biological processes via differentially expressed genes is limited when attempting to infer dysregulated pathways.
Highlights
In network biology, defining causal relationships among nodes is crucial for static and dynamic analysis [1,2]
It is known that gene expression or transcriptome refers to “what appears to happen in a biological system”, while the signaling network explicates “what makes it happen and what has happened in a complex view of the system” [11]
The ratio of eligible edges in the OmniPath edge list was higher than Kyoto Encyclopedia of Genes and Genomes (KEGG) based on both Genomics of Drug Sensitivity in Cancer (GDSC) and Gene Expression Omnibus (GEO) databases
Summary
In network biology, defining causal relationships among nodes is crucial for static and dynamic analysis [1,2]. The most available high-throughput data for inferring molecular relationships are arguably whole-transcriptome expression profiles analyzed with statistical models [3]. The challenge is extrapolating causality in signaling and regulatory mechanisms from a significant correlation between any given gene pair. Reverse engineering algorithms have been developed to tackle this challenge and to infer gene networks and regulatory interactions from expression profiles [5,6,7]. Inference of signaling networks can be directly inferred from (Phospho) proteomic and protein–protein interaction data [10]. These kinds of data are expensive and laborious to acquire. This, begs the question of whether gene expression profiles amplify the mechanism of signaling circuits, i.e., activatory/inhibitory relationships
Sign-in/Register to view highlights & summary Sign-in/Register to access full text options 
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a callDisclaimer: All third-party content on this website/platform is and will remain the property of their respective owners and is provided on "as is" basis without any warranties, express or implied. Use of third-party content does not indicate any affiliation, sponsorship with or endorsement by them. Any references to third-party content is to identify the corresponding services and shall be considered fair use under The CopyrightLaw.
