Integrative identification of key genes governing wilt resistance in using machine learning and WGCNA.

Introduction: wilt, caused by , is one of the most devastating diseases affecting global cotton () production. Given the limited effectiveness of chemical control measures and the polygenic nature of resistance, elucidating the key genetic determinants is imperative for the development of resistant cultivars. In this study, we aimed to dissect the temporal transcriptional dynamics and regulatory mechanisms underlying response to infection.

Methods: We employed a time-course RNA-Seq approach using the susceptible upland cotton cultivar Jimian 11 to profile transcriptomic responses in root and leaf tissues post- inoculation. Differentially expressed genes (DEGs) were identified, followed by weighted gene co-expression network analysis (WGCNA). To prioritize key candidate genes, we applied machine learning algorithms including LASSO, Random Forest, and Support Vector Machine (SVM).

Results And Discussion: A robust set of core genes involved in pathogen recognition (), calcium signaling (), hormone response, and secondary metabolism () were identified. Our findings provide novel insights into the spatiotemporal regulation of immune responses in cotton and offer valuable candidate genes for molecular breeding of wilt resistance.