Fig. 1: Study design.
From: Aberrant gene expression prediction across human tissues
a We aimed to predict whether protein-coding genes are aberrantly underexpressed across 49 human tissues based on DNA and, optionally, RNA-seq data of clinically accessible tissues. Therefore, we created a benchmark for aberrant underexpression prediction by processing 11,215 RNA-seq samples from 633 individuals across 49 tissues from GTEx. This yielded 11,200 underexpression outliers out of 99 million assessable gene-sample pairs (0.01%). The violin plots are cartoons qualitatively describing the signal we aim to model. b Assessing various variant and tissue annotations, we found that predictions could be significantly improved by weighting variant effects with tissue-specific isoform proportions and incorporating the expression variability of a gene. Further integration of expression measurements from clinically accessible tissues led to another two-fold improvement. The precision versus recall curve is a cartoon depicting the main take-homes of the study. Created in BioRender. Hölzlwimmer, F. (2025) https://BioRender.com/n29t312.
