Many of you might be aware of the commonly quoted success rate of ~10% in drug development. One of the efficient ways to rise this number is to add human genetics insights into the pipeline, as illustrated in the below plot from a 'Views and News' article I contributed to @GENBiotechJrnl. Thanks to @KevinADavies for the invite. The commentary is on a recent work in @Nature by @cureffi, @mnelsonxy and colleagues on refining the impact of genetic evidence on clinical success (nature.com/articles/s41586-0…). The plot below visualizes the impact of human genetics on drug success rate using a real world data. Biotechnology Innovation Organization (BIO) has a nice report based on an analysis of drug programs in various stages of development during 2011-2020 (bio.org/clinical-development…). In the report, there is a key figure that summarizes the success rates across different transition phases. I used that as the real world data and show alongside how those rates will be look like if human genetics support is integrated into the pipeline. Link to full text of the Views and News: liebertpub.com/doi/10.1089/g… I guess, I should mention that there is a new analysis similar to Minikel et al.'s is out in medRxiv by @adamauton and colleagues from 23andMe (medrxiv.org/content/10.1101/…). This is perhaps a very important paper for 23andMe at this time, as it showcases the value of their data for drug development. Here, the authors use a much larger genetic association data based on self-reported phenotypes of ~15 million 23andMe participants and come to a similar conclusion as Minikel et al. But they also challenge Minikel et al.'s claim that effect size and minor allele frequency doesn't impact success rates. I'll discuss on that on another day, but for now I recommend reading both papers, and of course, my commentary :) liebertpub.com/doi/10.1089/g…