Fig. 1: Study Design with selection process of the genes.
From: Self-supervised attention-based deep learning for pan-cancer mutation prediction from histopathology
A, B Patient numbers for each tumor type in The Cancer Genome Atlas (TCGA) and the Clinical Proteomic Tumor Analysis Consortium (CPTAC) dataset. C Flowchart showing the preprocessing steps for the training and validation cohort. Furthermore, an outline of the gene selection process. D Overview Area under the receiving operating curve (AUROC) results for internal cross-validation in TCGA. E Overview AUROC results for external validation on CPTAC. The plots are based on the original AUROC values with 5 decimal digits, while numbers in the manuscript text have been rounded to two decimal digits. (Icons were used from Servier Medical Art provided by Servier, licensed under a Creative Commons Attribution 3.0 unported license).
