An integrative computational approach to predict stress-specific candidate and shared genes in multiple plant stresses

Abstract

Plants are under the influence of various stresses that negatively impact their growth and development. Despite vast understanding of stress-related pathways and genes, significant success in developing stress-resistant crops is not achieved yet. At molecular-level, this is attributed to incomplete or partial understanding of regulatory interactions among key pathway genes that might provide new insights into the molecular pathways/mechanisms, thus could help developing new stress-resistant plant varieties. Therefore, in this work, by taking into account the interactions among stress-related genes, an integrated computational pipe-line was implemented to predict the most potential ‘key genes’ in model plant Arabidopsis thaliana during the three most common abiotic stress conditions—salt, heat and cold. Overall, the sequential gene selection approach is comprised of (i) differential expression studies among stress and control samples (ii) inferring stress-related gene co-expression networks, and (iii) logistic regression-based gene prioritization. During the analyses, various key candidate genes were predicted in salt, heat and cold stress including a significant number of cross-talking genes among stresses. Comprehensive analyses also provided various systems-level insights into the topological characteristics of stress-related networks. Gene ontology enrichment analyses also indicated potential roles of identified candidate genes in stress. Applicability of this approach in selecting the key stress-related genes was also demonstrated in tomato (Solanum lycopersicum) that identified gene Solyc01g097190 to be coordinating among salt, heat and cold stress; this yet uncharacterised Arabidopsis homolog gene might provide ample scope for exploring its roles in multiple stresses. Overall, this computational approach could be advanced (by including other omics data) and implemented to predict candidate stress-related genes in other crops or economically important plants as well.