Fig. 3: DeepNull improves phenotype prediction compared to Baseline.
From: DeepNull models non-linear covariate effects to improve phenotypic prediction and association power
The X-axis provides the phenotype names and the Y-axis is the R2 where R is Pearson’s correlation between true and predicted value of phenotypes. Center of each bar indicates the computed R2 over all samples and the error bars indicate the standard error. Standard errors are computed by performing bootstrapping for each phenotype (n = 1000 bootstrapping trials). Phenotypic abbreviations: alkaline phosphatase (ALP), alanine aminotransferase (ALT), aspartate aminotransferase (AST), apolipoprotein B (ApoB), glaucoma referral probability (GRP), LDL cholesterol (LDL), sex hormone-binding globulin (SHBG), and triglycerides (TG).
