> In the land of drug development [...] Fine, but this is the land of genome-wide association studies. I am unaware of any overlap, considering that such 'GWAS

> In the land of drug development [...] Fine, but this is the land of genome-wide association studies. I am unaware of any overlap, considering that such 'GWAS' studies require tens or hundreds of thousands of participants to get any definitive signal... Such work is the nature of statistical genetics. > So I question whether they might be fitting a model to noise here. On what basis?