A large-scale multi-omics polygenic risk score analysis identified candidate risk locus associated with rheumatoid arthritis

Jingni Hui; Dan He; Chen Liu; Panxing Shi; Ruixue Zhou; Meijuan Kang; Ye Liu; Yifan Gou; Bingyi Wang; Shiqiang Cheng; Xuena Yang; Chuyu Pan; Wenming Wei; Feng Zhang

doi:10.1016/j.jbspin.2024.105841

A large-scale multi-omics polygenic risk score analysis identified candidate risk locus associated with rheumatoid arthritis

Joint Bone Spine. 2024 Dec 26:105841. doi: 10.1016/j.jbspin.2024.105841. Online ahead of print.

Authors

Jingni Hui¹, Dan He¹, Chen Liu¹, Panxing Shi¹, Ruixue Zhou¹, Meijuan Kang¹, Ye Liu¹, Yifan Gou¹, Bingyi Wang¹, Shiqiang Cheng¹, Xuena Yang¹, Chuyu Pan¹, Wenming Wei¹, Feng Zhang²

Affiliations

¹ Key Laboratory of Trace Elements and Endemic Diseases of National Health and Family Planning Commission, School of Public Health, Health Science Center, Xi'an Jiaotong University, No. 76 Yan Ta West Road, 710061 Xi'an, China.
² Key Laboratory of Trace Elements and Endemic Diseases of National Health and Family Planning Commission, School of Public Health, Health Science Center, Xi'an Jiaotong University, No. 76 Yan Ta West Road, 710061 Xi'an, China. Electronic address: [email protected].

PMID: 39732430
DOI: 10.1016/j.jbspin.2024.105841

Abstract

Objective: This study aimed to investigate the associations of multi-omics polygenic risk score (PRS) and rheumatoid arthritis (RA) to identify potential genes/proteins and biological pathways.

Methods: Based on multi-omics data from 48,813 participants in the INTERVAL cohort, we calculated multi-omics PRS for 13,646 mRNAs (RNASeq), 308 proteins (Olink), 2,380 proteins (SomaScan), 726 metabolites (Metabolon), and 141 metabolites (Nightingale). Using the generalized linear model, we first evaluated the associations between multi-omics PRS and RA in 58,813 UK Biobank participants. The Gene Ontology (GO) project and Kyoto Encyclopedia of Genes and Genomes (KEGG) were performed to identify the functional pathways in RA. Furthermore, differential gene expression profile datasets were used to validate the identified genes/proteins in our study.

Results: We identified 59 transcriptomics PRS and 29 proteomics PRS significantly associated with RA. Both proteomics and transcriptomic PRS identified HLA-DQA2 (RNASeq: OR=1.19, P=1.18×10^-24; SomaScan: OR=1.24, P=4.43×10^-27) and AGER (RNASeq: OR=0.91, P=4.18×10^-4; SomaScan: OR=0.93, P=3.97×10^-3) were significantly associated with RA. Proteomic PRS from different profiling platforms (SomaScan and Olink) identified a consistent association between TFF3 (SomaScan: OR=0.90, P=4.08×10^-6; Olink: OR=0.93, P=4.87×10^-3) and RA. The identified gene/proteins were mainly enriched in the NF-kappa B signaling pathway (hsa04064, P=5.06×10^-5) and Cytokine-cytokine receptor interaction (hsa04060, P=2.49×10^-4). In addition, a total of 12 candidate genes in our study were verified in two independent GEO datasets, such as FLOT1 and ABCF1.

Conclusion: Our findings provide novel insights into the involvement of identified genes/proteins and pathways in the pathogenesis of RA from multi-omics levels.

Keywords: Rheumatoid arthritis (RA); biological pathways; multi-omics levels; polygenic risk score (PRS).