Table 2.
Association results for the rs1712779 and rs10831496 in the Huoshenshan and Union Cohorts in the main analyses.
| SNPs | Chr. (cytoband) | Cohorts | Minor allele frequencies | ORs | P values | Pheterogeneity | |
|---|---|---|---|---|---|---|---|
| (Minor/major allele) | Casesa | Controlsb | (95% CIs) | ||||
| rs1712779 (T/A) | 11q23.3 | Huoshenshan | 0.100 | 0.166 | 0.52 (0.38–0.70) | 1.58 × 10−5 | 0.53 |
| Union | 0.082 | 0.166 | 0.44 (0.28–0.67) | 1.85 × 10−4 | |||
| Meta (in-house data) | 0.096 | 0.166 | 0.49 (0.38–0.63) | 1.38 × 10−8 | |||
| Wang et al. study | 0.081 | 0.113 | 0.31 (0.11–0.89) | 0.029 | 0.092 | ||
| Erola et al. study (Asian) | NA | NA | 0.78 (0.62–0.98) | 0.031 | |||
| Meta (external data) | NA | NA | 0.75 (0.60–0.93) | 0.010 | |||
| Meta (all) | NA | NA | 0.62 (0.52–0.73) | 1.36 × 10−8 | 0.010 | ||
| rs10831496 (A/G) | 11q14.2 | Huoshenshan | 0.279 | 0.194 | 1.61 (1.29–2.00) | 2.04 × 10−5 | 0.63 |
| Union | 0.295 | 0.191 | 1.77 (1.28–2.43) | 4.75 × 10−4 | |||
| Meta (in-house data) | 0.283 | 0.193 | 1.66 (1.38–1.98) | 4.04 × 10−8 | |||
| Wang et al. study | 0.198 | 0.231 | 0.77 (0.42–1.42) | 0.40 | 0.22 | ||
| Erola et al. study (Asian) | NA | NA | 1.15 (0.93–1.43) | 0.20 | |||
| Meta (external data) | NA | NA | 1.10 (0.90–1.35) | 0.36 | |||
| Meta (all) | NA | NA | 1.38 (1.21–1.58) | 2.31 × 10−6 | 0.0063 | ||
Chr. chromosome, CI confidence interval, OR odds ratio, SNP single-nucleotide polymorphism.
aThe severe and critical types of COVID-19 patients were defined as the “cases”.
bThe mild and moderate types of COVID-19 patients were defined as the “controls”.
After quality controls, the Huoshenshan cohort includes 663 cases and 322 controls, and the Union cohort includes 200 cases and 207 controls. In the main analyses, the associations in each cohort were calculated using logistic regression models adjusted for age, gender, comorbidities (including hypertension, type 2 diabetes, and coronary artery diseases) and the top five principal components. Meta-analyses of the association results generated from the Huoshenshan and Union cohorts were conducted using a fixed-effect model. ORs and 95% CIs are shown with respect to the minor alleles. The index rs1712779 at Chr. 11q23.3 was imputed with high confidence (IMPUTE4 info scores were 0.97 in the Huoshenshan cohort, and 0.98 in the Union cohort). The rs10831496 at Chr. 11q14.2 was genotyped by Affymetrix Axiom® World Arrays. The replication results presented here included the results of two other COVID-19 GWASs: (1) Wang et al. study (PMID: 33298875). This study contained 69 severe/critical COVID-19 patients (cases) and 215 mild/moderate COVID-19 patients (controls) from China, who were subjected to whole-genome sequencing. We only retained the subjects from Hubei Province, and achieved 43 cases and 127 controls. To increase power, 200 naïve controls from Hubei Province, as well as 662 genetically well-matched naïve controls according to CONVERGE (China, Oxford and Virginia Commonwealth University Experimental Research on Genetic Epidemiology) consortium (Changsha, n = 72; Chongqing, n = 295; Nanchang, n = 295) were also included. These naïve controls were subjected to whole-genome sequencing. Thus, 43 cases and 989 controls were used in the final association test. The association test was performed in logistic regression models adjusting for age and gender. (2) Erola et al. study (PMID: 33307546). This study included the populations of European ancestry (primary analyses), East Asian, South Asian, and African ancestries. Here we showed the results from the populations of the Asian ancestries (meta-analysis of East Asian ancestry and South Asian ancestry; 386 critically ill COVID-19 patients and 1930 naive controls).