Limitations and Cautions for the Interpretation of Genome Wide Association Studies and Polygenic Scores: Lessons from Lewontin
Though written almost a half-century before the age of massive genome-wide association studies (GWAS) and the related development of polygeneic scores (GPS), several elements of Lewontin’s work remain relevant to thinking about the limitations of GWAS / PGS for discovering causal relationships. Lewontin’s work on population structure points towards why it is so difficult to successfully address problems of population stratification / cryptic population structures in such research. His work on gene-environment interactions point towards why we should not expect associations found in one population to be predictive in others. And his work on the ways in which organisms construct their environments points towards the difficulty of using GWAS / PGS as an entry into developmental biology.