New paper "Trait selection strategy in multi-trait GWAS: Boosting SNPs discoverability" has been published! https://www.cell.com/hgg-advances/fulltext/S2666-2477(24)00058-7
For the user of multi-trait GWAS, a question to think about when using the method is which traits to analyze together. As we show in this paper, the performance of multi-trait GWAS (i.e. the number of new associations) depends on the combination of traits.
Thus, we addressed the question of how should we select traits to be analyzed by multi-trait GWAS so that the performance improves. An interesting take-away message is that a common strategy (i.e. selecting clinically similar traits) does not particularly do well. Our model analysis provides the features that impact the performance of multi-trait GWAS, which can be leveraged in future studies to improve detecting associations in complex traits.
