Transcriptomic profiling reveals immune signatures associated with potential COVID-19 susceptibility and a predictive framework in a Chinese cohort.

Introduction: Despite significant interindividual susceptibility to COVID-19, the molecular basis of host vulnerability in the Chinese population remains poorly defined. Methods: Leveraging the large-scale Omicron exposure following China's transition from "zero-COVID," we conducted transcriptomic profiling of peripheral blood mononuclear cells (PBMCs) from 173 volunteers stratified into high susceptibility (HS) and low susceptibility (LS) groups based on SARS-CoV-2-specific antibody levels and retrospective symptom assessments. Differential expression analysis, protein-protein interaction (PPI) network analysis, weighted gene co-expression network analysis (WGCNA), and single-sample gene set enrichment analysis (ssGSEA) were performed. An ensemble machine-learning model was trained on 80% of the cohort and evaluated on the remaining 20% test set. Results: Differentially expressed genes were enriched in neutrophil recruitment and T cell activation, while PPI analysis prioritized hub genes involved in inflammation, chemotaxis, and endothelial integrity. WGCNA identified two modules negatively correlated with high susceptibility, enriched in MHC class II antigen presentation and T cell activation/epigenetic regulation, respectively. HS individuals showed enrichment of interferon-related transcriptional signatures, together with reduced expression of interferon receptor-associated genes and suppression of B-cell receptor and complement pathways. The ensemble model achieved an AUC of 0.92 on the independent test set. Discussion: These findings identify transcriptomic signatures associated with potential COVID-19 susceptibility across innate, adaptive, and vascular immune-related axes and provide an exploratory framework for risk classification and potential precision prevention.