Computational genomics insights into cold acclimation in wheat

Abstract

Development of cold acclimation in crops involves transcriptomic reprograming, metabolic shift, and physiological changes. Cold responses in transcriptome and lipid metabolism has been examined in separate studies for various crops. In this study, integrated computational approaches was employed to investigate the transcriptomics and lipidomics data associated with cold acclimation and vernalization in four wheat genotypes of distinct cold tolerance. Differential expression was investigated between cold treated and control samples and between the winter-habit and spring-habit wheat genotypes. Collectively, 12,676 differentially expressed genes (DEGs) were identified. Principal component analysis of these DEGs indicated that the first, second, and third principal components (PC1, PC2, and PC3) explained the variance in cold treatment, vernalization and cold hardiness, respectively. Differential expression feature extraction (DEFE) analysis revealed that the winter-habit wheat genotype Norstar had high number of unique DEGs (1884 up and 672 down) and 63 winter-habit genes, which were clearly distinctive from the 64 spring-habit genes based on PC1, PC2 and PC3. Correlation analysis revealed 64 cold hardy genes and 39 anti-hardy genes. Cold acclimation encompasses a wide spectrum of biological processes and the involved genes work cohesively as revealed through network propagation and collective association strength of local subnetworks. Integration of transcriptomics and lipidomics data revealed that the winter-habit genes, such as COR413-TM1, CIPKs and MYB20, together with the phosphatidylglycerol lipids, PG(34:3) and PG(36:6), played a pivotal role in cold acclimation and coordinated cohesively associated subnetworks to confer cold tolerance.