Integrative multi-omics framework for causal gene discovery in Long COVID.

Area of Science:

• Genetics
• Immunology
• Computational Biology

Background:

• Long COVID (Post-Acute Sequelae of SARS-CoV-2 infection) affects 10-20% of COVID-19 survivors with persistent multisystem symptoms.
• Genetic factors contributing to Long COVID risk are not well understood, despite known demographic and clinical risk factors.

Purpose of the Study:

• To identify causal genes and network drivers of Long COVID using an integrated multi-omics approach.
• To elucidate the genetic architecture and heterogeneity of Long COVID.
• To discover potential therapeutic targets for Long COVID.

Main Methods:

• Developed a multi-omics framework integrating Transcriptome-Wide Mendelian Randomization (TWMR), Control Theory (CT), eQTL, GWAS, RNA-seq, and Protein-Protein Interaction (PPI) networks.
• Prioritized candidate genes involved in viral response, carcinogenesis, immune regulation, and cell cycle control.
• Performed enrichment analyses and identified Long COVID subtypes based on causal gene expression.

Main Results:

• Identified 32 candidate genes, including 13 novel, implicated in Long COVID pathogenesis.
• Revealed shared genetic architecture with syndromic, metabolic, autoimmune, and connective tissue disorders.
• Discovered three distinct symptom-based Long COVID subtypes, highlighting disease heterogeneity.
• Developed an open-source Shiny application for data exploration.

Conclusions:

• The integrative framework identified novel causal mechanisms and therapeutic targets for Long COVID.
• Findings advance precision medicine strategies for Long COVID by characterizing subtypes and underlying genetic drivers.
• Understanding genetic underpinnings is crucial for developing targeted Long COVID treatments.