Deep learning predicts DNA methylation regulatory variants in the human brain and elucidates the genetics of psychiatric disorders.

For example, the enrichment was only nominally significant for schizophrenia (fold = 2.1, P = 0.032), neuroticism (fold = 2, P = 0.030), and education years (fold = 1.8, P = 0.034), whereas the top 10% ranked variants from the brain-specific INTERACT showed stronger evidence for the same traits (schizophrenia, fold = 2.5, P = 1.5 × 10 −4 ; neuroticism, fold = 2.5, P = 3.6 × 10 −4 ; education years, fold = 2.3, P = 2.3 × 10 −4 ).

also in 7,732 other papers

There was a clear trend that variants of higher LD with the top mQTLs showed stronger association evidence, but predicted effects of variants were not associated with their LD strength with the top variant of the highest effect ( Fig. 2 C ).

also in 19,460 other papers