Deconvoluting complex correlates of COVID-19 severity with a multi-omic pandemic tracking strategy

Abstract

The SARS-CoV-2 pandemic has differentially impacted populations across race and ethnicity. A multi-omic approach represents a powerful tool to examine risk across multi-ancestry genomes. We leverage a pandemic tracking strategy in which we sequence viral and host genomes and transcriptomes from nasopharyngeal swabs of 1049 individuals (736 SARS-CoV-2 positive and 313 SARS-CoV-2 negative) and integrate them with digital phenotypes from electronic health records from a diverse catchment area in Northern California. Genome-wide association disaggregated by admixture mapping reveals novel COVID-19-severity-associated regions containing previously reported markers of neurologic, pulmonary and viral disease susceptibility. Phylodynamic tracking of consensus viral genomes reveals no association with disease severity or inferred ancestry. Summary data from multiomic investigation reveals metagenomic and HLA associations with severe COVID-19. The wealth of data available from residual nasopharyngeal swabs in combination with clinical data abstracted automatically at scale highlights a powerful strategy for pandemic tracking, and reveals distinct epidemiologic, genetic, and biological associations for those at the highest risk.

Fig. 1. SARS-CoV-2 pandemic tracking from residual NP swabs and abstracted EHR data combined with genetic ancestry inference allows identification of high risk populations and examination of its interaction with viral phylogeny and disease severity.

A We collected samples from 736 SARS-CoV2 positive and 313 negative patients between Mar-Aug 2020 with clinical severity scores ranging from 1 (ambulatory) to 8 (death). B Examples of individual patient trajectories in COVID-19 severity score as abstracted from the electronic healthcare record. C Severity scores abstracted directly from the electronic health record daily for thirty days before and after the positive NP swab test on all included patients with severity score ≥ 4 (hospitalized, needs oxygen) demonstrates significant variability in patient course. D Whole genome sequencing from DNA isolated from 150 ul of NP swab VTM yielded sequence on >95% of samples with mean of means coverage 2.6X. E RNA sequencing using shotgun sequencing recovered consensus SARS-CoV-2 sequence on the majority of NP swabs with a clinical PCR CT value <30. ARTIC primer enrichment increased this yield (Supplementary Fig. 1D). F Genetic ancestry admixture of individuals with positive versus negative COVID-19 tests in the present study. Individuals with Indigenous American ancestry are overrepresented in cases, whereas controls show more European and South Asian genetic ancestry. G Self-reported (top) and genetic ancestry (bottom) of enrolled COVID-19 + individuals over time reveals disproportionate representation of Hispanic/Latino ethnicity and Indigenous American ancestries during summer pandemic wave, whereas the first wave is seen to have predominantly affected non-Hispanic individuals and individuals of European genetic ancestry. H Phylogenetic reconstruction of SARS-CoV-2 sequences. Tip colors correspond to the inferred genetic ancestry of the infected hosts, whose consensus SARS-CoV-2 sequences were isolated and used for inferring the viral phylogeny. Horizontal lines to the right of the phylogeny indicate host severity scores corresponding to the tips of the phylogeny. Severity score codes are displayed in Supplementary Table 1.

Fig. 2. COVID-19 severity is associated with local-ancestry-specific risk loci via admixture mapping, and is also correlated with metagenomic features of the NP transcriptome.

A Ancestry-specific risk loci found in African and Oceanian ancestries, respectively after correcting for overall genetic ancestry proportion, BMI, sex, and age. Each colored dot represents a window of the genome. Black lines represent ancestry-specific thresholds determined by the method of Shriner et al. Thresholds determined by running one thousand association tests on random permutations of case-control labels are displayed in Figure S5. B Traits associated with genomic regions statistically enriched for disease severity in the GWAS catalog. For additional information including a full list of previously reported SNPs and neighboring genes, see Supplementary Data 1. All summary statistics are available at covid-omics.org. C Schematic of multiomic pandemic tracking strategy. Created with BioRender.com. D Uniform manifold approximation and projection (UMAP) of patient Nasal Microbiome abundances colored by patient COVID-19 severity score. (E) Regression of species-specific abundance against continuous disease severity, corrected for age, sex and BMI, identified P. yeei abundance in the nasopharyngeal microbiome as associated with high severity COVID-19 infections (Bonferroni adjusted p = 7e−04 (two-sided)).