Overview of the OmicsGWAS workflow for genetic correlation, genetic pleiotropy, and multi-omics integration analysis based on GWAS summary statistics (IMAGE)
Caption
The pipeline takes GWAS summary statistics from two or more traits as input. The omics_format_data module first performs quality control, format standardization, and allele harmonization to ensure cross-trait and cross-population consistency. For dual-trait analyses, the workflow proceeds through several sequential modules: (1) Genetic correlation analysis, including global genetic correlation (LDSC, HDL, GNOVA) and local genetic correlation (LAVA, SUPERGNOVA, HDL-L, HESS, GWAS-PW); (2) Genetic pleiotropy analysis, integrating MTAG, CPASSOC, PLACO, pleioFDR, MiXeR, and CCGWAS to identify shared genetic loci and pleiotropic architecture, with significant results from MTAG and CPASSOC further filtered via the CPASSOC_mtag_sig module; (3) Gene prioritization, employing FUMA, MAGMA, POPS, fine-mapping, XWAS, and fastBAT-GCTA for candidate gene annotation and signal refinement; and (4) Downstream functional interpretation, leveraging FLAMES, FOCUS, FUSION, UTMOST, MetaXcan, xQTL-GWAS SMR/MR/COLOC, multi-omic MR/SMR, and miRNA-xQTL MR to systematically prioritize causal genes and molecular pathways through expression regulation, tissue-specific transcriptomic association, colocalization, and causal inference. Beyond dual-trait analysis, the framework supports multi-GWAS extensions: METAL for cross-population meta-analysis of the same trait, and GenomicSEM for modeling shared genetic factors across multiple traits. The get_full_* functions (denoted by asterisks) enable direct retrieval of database-level xQTL resources, including MacroMap, pancell, dynamic CD4, ImmuNexUT, OASIS, tenk10k, and eQTL Catalogue.
Credit
Xu zhang，Kailai Li, Yasi Zhang, Cenwen Liu, Shunyi Huang, Yingzhao Jin, Yugeng Gao, Quan Cheng，Yunyun Mei
Usage Restrictions
none
License
Original content