The recent perspective paper in Cell An Expanded View of Complex Traits: From Polygenic to Omnigenic [Pubmed] [Doi] by E.A. Boyle, Y. Li, and J.K. Pritchar [Cell, 15 June 2017] has sparked quite a number of discussions, as indicated by its high Altmetric score (366 at the time I am writing this).
The authors conclude that “.. many complex traits are driven by enormously large numbers of variants of small effects, potentially implicating most regulatory variants that are active in disease-relevant tissues.”
To explain these observations, they “propose that disease risk is largely driven by genes with no direct relevance to disease and is propagated through regulatory networks to a much smaller number of core genes with direct effects. If this model is correct, then it implies that detailed mapping of cell-specific regulatory networks will be an essential task for fully understanding human disease biology.”
For some reasons some commentators see in this paper New concerns raised over value of genome-wide disease studies, as Nature reporter Ewen Callaway entitles his piece [Nature, 15 June 2017], and others ask: The GWAS hoax….or was it a hoax? Is it a hoax?, as blogger Ken Weiss [Bog post, 17 June 2017].
Personally, I cannot subscribe to these views. GWAS are not a failure, experiments to be concerned about, or a hoax!
If you remember where physics stood at the turn of the 20th century: we believed that all fundamental laws had been discovered, and only tiny details, such as the speed at which the Earth moves through the aether, needed to be clarified. Then came the “failure” of the Michelson-Morley experiment, which could not determine the Earth’s movement through the aether. This “failed” experiment toppled the classical view of physics, and the aether, and it took scientists like Einstein to invent radically new theories to explain the observed data.
I believe that GWAS are the Michelson-Morley experiment of biology – their results question the classical view of genetics, e.g. that “one might expect disease-causing variants to cluster into key pathways that drive disease etiology” [Boyle et al.].
The paper of Boyle et al. does not more and not less than pointing this out, and it offers some new directions on where to go next. We should avoid making this a discussion about whether GWAS were worth the effort – they clearly are as valuable to biology as was the Michelson-Morley experiment to physics.
In terms of future directions, and disclosing my personal interest here, I like to add that the one thing that I missed in the Boyle et al. paper in terms of future directions is the role that GWAS with intermediate traits will play, such as GWAS with metabolomics and proteomics.
These multi-omics GWAS carry an immense load of biological information, which can help explaining many of the currently “spurious” disease associations, and are totally under-explored at the moment.
Read my posts on metabolomics GWAS, proteomics GWAS, and also metabolomics EWAS to learn more about these genetic treasure troves.