HIGHLIGHTS
SUMMARY
Presumably, then, with sufficient association data and a diverse and dense collection of functional annotations across the genome, it might be possible to train a model to predict loci - and perhaps even variants - that relate to specific sets of traits relative to all others. The authors then applied a penalized binomial logistic regression model (LASSO) to each trait group using publicly available genome wide features such as DNase hypersensitivity and histone or transcription factor ChIP-seq (Supplementary Table 2), along with gene_expression, phastCon score, and size of locus. To assess why these 6 traits . . .
If you want to have access to all the content you need to log in!
Thanks :)
If you don't have an account, you can create one here.