Integrated multi-sample transcriptomic analysis of COVID-19 patients against controls using a bioinformatics pipeline

Prior coronavirus disease 2019 (COVID-19) transcriptomic studies using diverse methods for differential gene expression (DGE) profiling of specific samples yielded inconsistent results. To validate the shared molecular patterns of COVID-19 across cell, tissue, and systemic levels, we conducted a systematic rank combination meta-analysis of differentially expressed gene (DEG) profiles sourced from various sample types using a standardised bioinformatics pipeline consisting of DESeq2, RankProd, and weighted gene correlation network analysis (WGCNA). Consistently upregulated ISGs (including key hub gene IFIT2), compared with interleukins were identified in swab samples, reflecting dominant innate immune responses at the viral entry point. Blood samples revealed diverse gene functions in immune and neurological regulation, highlighting the complex interplay of systemic regulation. Significant enrichment of immunoglobulin-related and extracellular matrix genes indicates their role in the host adaptive immunity and long-term host responses in tissue samples. Novel key hub genes in tissue samples, GPD1 and CYP4A11 related to metabolic dysregulation were identified, potentially contributing to the severity of the disease. These findings portray the molecular basis of COVID-19 progression from localised innate responses to systemic effects and finally tissue-specific adaptive immunity and remodelling, providing insights that may inform diagnostic and therapeutic development.