Polygenic regression uncovers trait-relevant cellular contexts through pathway activation transformation of single-cell RNA sequencing data

Integration of human organoids single‐cell transcriptomic profiles and human genetics repurposes critical cell type‐specific drug targets for severe COVID‐19

“…To diminish the impact of this limitation, we adopted a powerful approach by incorporating a large-scale GWAS summary dataset and human organoids scRNA-seq data with a large amount of cells, as reference to previous studies. 7,18,19,35,58 Second, the identification of COVID-19-relevant cell types or subpopulations does not imply causality but may reflect indirect discovery of causal phenotype-cell associations, analogous to earlier studies. 19,58 Third, we removed the MHC region from all genomic analyses to reduce the influence of the complex genetic architecture and extensively high levels of LD, parallel to previous studies.…”

“…Multiple lines of evidence 18,19,35,58,59 have demonstrated that integrating scRNA-seq data and polygenic risk signals from GWAS is a promising approach to uncover the cellular mechanisms through which these variants drive complex diseases. In this study, we sought stimulate host innate antiviral responses.…”

“…To distinguish critical cell types/subpopulations by which genetic variants influence COVID‐19 severities, we implemented our own developed pathway‐based polygenic regression method, scPagwas (version 1.1.0), 35 to integrate GWAS summary data on three COVID‐19 outcomes with human organoids and fetal scRNA‐seq datasets. Initially, scPagwas annotates SNPs to their proximal genes (a default window size of 20 kb) of the corresponding pathways, which are based on the experimentally validated canonical pathways in the Kyoto Encyclopedia of Genes and Genomes (KEGG) database 36 .…”

“…The major histocompatibility complex region (chr6: 25–35 Mbp) is removed because of the extensive LD in this genomic region. For more detailed information, please refer to the original paper 35 …”

“…93 Brain organoids provide a promising tool for uncovering the pathophysiologic mechanisms and potential therapeutic options for neuropsychiatric complications of severe COVID-19. 94 We leveraged the scPagwas method 35 to integrate the GWAS summary dataset on very severe COVID-19 and human cerebral organoids scRNA-seq data. 57 Among eight main cell types, we identified that both endothelial cells (p = 6.96 Â 10 À6 ) and microglia (p = 5.29 Â 10 À5 ) yielding higher TRSs were significantly associated with very severe COVID-19 compared with other cell populations (Figure 7A-C), recalling that these two cell types were identified to be associated with COVID-19 severities in human fetal brain tissue (Figure S6).…”