NTR), and adipose (METSIM) SNP weight sets (https://www.dropbox.com/s/ep3dzlqnp7p8e5j/focus.db?dl=0), with genes discovered using the PsychENCODE weights finemapped specifically using that panel given the different LD parameters and its more complete set of genes with cis-heritable models in that one tissue. The multi-tissue finemapping panel contains several other non-brain tissues, and thus, some GReX models that would not have been available in brain and blood. We sought to balance maximising the number of models available for finemapping, whilst acknowledging that some of the tissues in this panel are less likely to be disease relevant. As FOCUS allows the null mode that the causal feature is not typed to be predicted as a possible member of the credible set, we excluded any genes for which that occurred. The credible set was defined by summing normalised PIP such that ρ was exceeded, sorting the genes, and then including those genes until at least ρ of the normalised-posterior mass is explained, as described in more detail elsewhere (Mancuso et al., 2019; Reay et al., 2021). As an exploratory analysis, we also applied finemapping to the PWAS and spliceWAS results, although the limited number of SNP weights available for protein and alternative splicing means that the
