A Bayesian Framework to Account for Complex Non-Genetic Factors in Gene Expression Levels Greatly Increases Power in eQTL Studies

@ekernf01 @jjloverp @SciReports in PEER and QTL studies this is just treated as a "hidden factor matrix", would be useful to have a more precise definition https://t.co/sOI0HyTB9Z

@pinkddle Would you count SCVI or PEER as automated variational inference? https://t.co/sZhKKEv4Lc https://t.co/VWdg375uZm PEER has the OG John Winn as a coauthor!

@nafiz_h An example of this approach is correcting for hidden factors in gene expression when regressing against genotype https://t.co/VbZHJSBnbe . More recently prob programming was able to infer causal estimates in large genome association studies https:

RT @MarkGerstein: A Bayesian Framework to Account for Complex Non-Genetic Factors in Gene Expression Levels Greatly Increases Power in #eQT…