To summarise, three statistical genetics data analyses are carried out as outlined in Figure 1C. Lead variant annotation, and lead variant to tag variant expansion methods, which include fine-mapping/credible set analysis and linkage-disequilibrium expansion are described in full here: https://genetics-docs.opentargets.org/our-approach/assigning-traits-to-loci. Disease-molecular trait colocalisation analysis for studies with full summary statistics is explained in more detail here: https://genetics-docs.opentargets.org/our-approach/colocalisation-analysis. Similar analyses have been conducted for GWAS studies without full summary statistics, using an approximate colocalisation heuristic based on variant probabilities from the PICS method. These analyses then feed into the causal inference L2G analysis pipeline to connect the associated loci to genes, utilising underlying evidence in order to ultimately rank the genes most likely to be underlying the associated trait/disease (Figure 1C).
