We’ve posted an updated version of our bioRxiv preprint on the ability of genome-wide association studies to spur subsequent research into newly associated genes. Most importantly, we found that the effect of GWAS has declined dramatically, suggesting that researchers are following-up on GWAS much less than they used to. Although our work does not show why recent GWAS have much less impact than earlier GWAS, we hope it motivates efforts to encourage GWAS follow-up.
Our recent work with Erik Knudsen on variant calling in tumors is now on bioRxiv. The paper introduces a new approach for analyzing genetic variants called from next-gen sequencing of tumors. A key challenge in such sequencing is spurious variants calls, particularly when sequencing tumors that have been xenografted into mice. We introduce and characterize a simple BLAST-based algorithm for removing spurious calls caused by mouse contamination or paralogs in the genome. This algorithm is as effective as much more computationally or bioinformatically intense approaches. Our testing also revealed biases that may be introduced by commonly used variant callers, an important caution for the community. Congratulations Brian!
Our collaborative paper with Michael Hammer on polygenic adaptation in Siberians has been published in MBE. In the paper, we develop a comprehensive model of the join demographic history of Europeans, East Asians, and Siberians. We then use that model as a null model to scan for gene sets enriched in signatures of population-specific adaptation. Three of our hit gene sets are related to diet, particularly fat metabolism, consistent with adaptation to a fat-rich animal diet. Congratulations Benson! (Also thank you to co-author Ludmila Osipova for the accompanying photo.)