Whole-Exome Sequencing Explores Host Genetic Susceptibility of COVID-19 Severity in Unvaccinated Vietnamese Patients Infected with the Delta Variant: A Hypothesis-Generating Study

Background/Objectives: Host genetic variation contributes to the heterogeneous clinical outcomes of coronavirus disease 2019 (COVID-19), yet evidence from Southeast Asian populations remains limited. Methods: We investigated host genetic factors associated with COVID-19 severity in Vietnamese patients infected with the SARS-CoV-2 Delta variant using whole-exome sequencing. A total of 48 unvaccinated patients were enrolled. During sample-level quality control, one sample was excluded because the genetically inferred sex was discordant with the information recorded in the clinical medical record, leaving 47 samples (23 severe/critical and 24 mild/asymptomatic) for downstream genetic analyses. Gene-based association analysis was performed using MAGMA, followed by functional enrichment and interaction network analyses with Metascape and GeneMANIA. Results: Population structure analysis showed that the study participants clustered closely with the East Asian (EAS) reference population. No individual variant reached the prespecified multiple-testing-adjusted threshold, and no gene-level association survived Benjamini–Hochberg correction. Using a nominal MAGMA gene-based p < 0.01 threshold solely for exploratory prioritization, 44 genes were selected for downstream functional analyses. Functional enrichment highlighted RNA processing and mRNA maturation, together with mitochondrial, metabolic, immune-regulatory, and cellular homeostasis pathways. Network topology analysis further identified several highly connected genes, including CPSF4, MRPS34, ATP5MF, BUD31, SNRPD3, and HDAC1. Conclusions: These hypothesis-generating findings are consistent with a potentially polygenic contribution to severe COVID-19 and provide an exploratory systems-level framework for understanding host genetic susceptibility in the Vietnamese population, while identifying biologically plausible candidate genes and pathways for future validation.