Integrative omics identifies novel biomarkers and therapeutic targets for the treatment and prevention of COVID-19 phenotypes

Abstract

The coronavirus disease 2019, commonly referred to as COVID-19, is a highly contagious disease caused by the severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) and has caused over 700 million infections worldwide. COVID-19 exhibits significant variability in symptoms, prognosis, and severity among individuals, ranging from asymptomatic cases to lethal outcomes. In addition to clinical and social factors, host genetics has been identified as a crucial contributor to this observed variability. Furthermore, despite the development of effective vaccines, there is still a need for drugs to treat the disease, especially for different stages of the disease progression. Therefore, it is crucial to understand the genetic mechanisms of COVID-19 and to identify reliable biomarkers and potential therapeutic targets for different stages of the disease.

Many genome-wide association studies (GWAS) have been conducted to identify single nucleotide variants which are associated with COVID-19 phenotypes. While these studies provide important insight into the genetic underpinnings of the disease, they are not sufficient to infer causality. Integrating information from GWAS and other omics-based studies is a powerful method to identify the biological pathways through which these genetic loci affect COVID-19 infection and progression. Moreover, doing so under the framework of Mendelian randomization is a well-established method of identifying causal genes and proteins for a disease.

In this study, information from omics-based studies is integrated to conduct a transcriptome- and proteome- wide mendelian randomization study to identify genes and proteins with expression levels causally associated with three COVID-19 phenotypes: severe COVID-19, hospitalized COVID-19, and SARS-CoV-2 infection. Independent genetic instruments which are strongly associated with gene expression levels in 49 different tissues and protein expression levels in blood plasma were identified from the largest available expression and protein quantitative trait loci studies respectively. Then, mendelian randomization was conducted to identify causal genes and proteins for each of the COVID-19 phenotypes. The FDR q-value was used to adjust for multiple testing and sensitivity analyses were conducted to ensure the robustness of results.

This study shows that integrative -omics and Mendelian randomization are powerful tools for the identification of causal genes and proteins for a disease outcome. This study identified several genes and proteins which are causally associated with three COVID-19 phenotypes. Many of these causal associations were identified for the first time. In the transcriptome-wide study, 64, 63, and 20 causal genes were identified for severe COVID-19, hospitalized COVID-19, and SARS-CoV-2 infection respectively. These genes were enriched in biological processes related to type I interferons, interferon-gamma inducible protein 10 production, and chemokine (C-X-C motif) ligand 2 production. In the proteome-wide study, 2, 8, and 2 causal proteins were identified for severe COVID-19, hospitalized COVID-19, and SARS-CoV-2 infection respectively. Four of these proteins, Kinesin light chain 1, MRVI1, Calcium-binding and coiled-coil domain-containing protein 2, and PEST proteolytic signal-containing nuclear protein were novel causal proteins for COVID-19. These findings provide valuable insights into the underlying mechanisms of COVID-19 and offer new opportunities for developing potential therapeutic targets or biomarkers for the treatment and prevention of COVID-19.