Integrative Genomic Analysis Highlights Potential Genetic Risk Factors for Covid-19

Abstract

RATIONALE: The genes that influence the pathophysiology of COVID-19 have yet to be identified. Association analysis has found genetic loci for COVID-191. We used integrative genomics (IG) to combine gene expression and proteomic information with COVID-19 susceptibility loci in order to identify candidate genes for this disease. METHODS: For these analyses we used the COVID-19 Host Genetics Initiative genome-wide association (GWA) meta-analysis version 4 (COVID-19 positive versus COVID-19 negative), the Lung eQTL study2 (n=1,038), eQTLGen3 study (n=31,784) and the INTERVAL4 study (n=3,301). We conducted two IG methods (Bayesian Colocalization [coloc] and Summary Based Mendelian Randomization) to link gene and protein expression in lung and blood tissues with COVID-19 susceptibility loci. We identified the most consistently colocalized gene and conducted a Mendelian Randomization (MR) to assess the causal association of its protein ('exposure') with COVID-19 susceptibility ('outcomes'). Significant MR was set as P<0.05. RESULTS: The expression of 6 genes in lung and 12 in blood colocalized with COVID-19 susceptibility loci. SMR results demonstrated that the expression levels of 6 genes in lung tissue and 5 in blood were associated with COVID-19. Out of the candidate genes identified, two (ABO and SLC6A20) were within previously identified loci (Figure 1). Based on the SMR we found that the expression of SLC6A20 in lung was associated with a higher risk of COVID-19. Novel discovered associations included ERMP1, FCER1G, and CA11, genes which have been previously linked with respiratory diseases (i.e.: asthma) and host immune responses (i.e.: neutrophil and eosinophil counts). COVID-19 susceptibility also colocalized with plasma protein levels of ABO. Based on MR, ABO demonstrated a significant causal association (P = 2.10 × 10-5) with the risk of COVID-19 with increased levels of this protein in plasma associated with an increased risk of COVID-19. The top variant in the MR test (rs505922) was in complete linkage disequilibrium with the variant responsible for the blood O genotype, conferring reduced risk. CONCLUSIONS: This multi-omics approach led to the discovery of novel genes associated with COVID-19. We found that the ABO protein is a causal risk factor for COVID-19, with blood group O being protective against COVID-19. REFERENCES: 1. Ellinghaus, D. et al. N. Engl. J. Med. (2020). 2. Hao, K. et al. PLoS Genet. (2012). 3. Ṽsa, U. et al. bioRxiv. (2018). 4. Sun, B. B. et al. Nature. (2018) .