Back

Identifying Key Hub Genes that Attribute Varying Host Responses: A Longitudinal RNA-seq Analysis of SARS-CoV-2 Delta and Omicron Infections.

Bajracharya, D.; Jansen, R.; Yan, C.; Ramamoorthy, S.

2025-06-24 bioinformatics
10.1101/2025.06.23.654375 bioRxiv
Show abstract

Coronavirus disease 2019 (COVID-19), caused by SARS-CoV-2, has severely impacted global health, with different Variants of Concern (VOCs) resulting in varied clinical outcomes. Among these, the Delta and Omicron variants have drawn significant attention due to their distinct characteristics--Delta is associated with higher virulence and Omicron has greater transmissibility but reduced severity. This study compares host immune responses to the Delta and Omicron variants using RNA-seq data from infected primary human airway epithelial cells. Both variants triggered a robust antiviral innate immune response by 2 days post-infection (dpi). However, Omicron was found to elicit a more rapid immune response, showing pathway enrichment at 1 dpi. In contrast, Delta displayed no immune-related pathway activation within the first 24 hours, suggesting it may evade early immune detection, promoting increased viral replication. By 3 dpi, Delta induced a more aggressive immune response, particularly in pathways related to cell death and pro-inflammatory signaling, such as "programmed cell death" and "regulation of cell death." Weighted gene co-expression network analysis revealed distinct immune-related genes: Delta infections were characterized by hub genes like MYD88 and IL1R1 involved in pro-inflammatory responses, HLA-A & B, NLRC5 and PSMB9 involved in antigen presentation and TNSF10 and IFR1 involved in pro-apoptotic processes. Conversely, Omicron infections were marked by hub genes such as CXCL1 and CXCL8 involved in immune cells recruitment, MET and LYN involved in reducing hyperinflammatory responses and maintaining immune balance, and IFI44, EIF2AK2 and IFIT5 responsible for the sensing of viral RNA among others.

Matching journals

The top 8 journals account for 50% of the predicted probability mass.

50% of probability mass above

"Similar papers" are the closest papers from that journal in the model's embedding space. They show what the match is built on, but the ranking comes mostly from a classifier over the whole training set, not from these examples alone.