Journal: NAR Genomics and Bioinformatics
Article Title: Cancer drug sensitivity estimation using modular deep Graph Neural Networks
doi: 10.1093/nargab/lqae043
Figure Lengend Snippet: Ablation study: performance measures ± standard error of the mean. ‘ \documentclass[12pt]{minimal} \usepackage{amsmath} \usepackage{wasysym} \usepackage{amsfonts} \usepackage{amssymb} \usepackage{amsbsy} \usepackage{upgreek} \usepackage{mathrsfs} \setlength{\oddsidemargin}{-69pt} \begin{document} $\checkmark$\end{document} ’ indicates included, ‘×’ indicates excluded model components; the first two rows are the complete CANDELA model with the two different pre-training stragies. Models marked ‘ † ’ are significantly worse than CANDELA using only one pre-training task. Models marked ‘*’ are significantly better (pairwise t -test, α = 0.05, corrected for multiple testing using Holm–Šídák). Bold values indicate the best configuration
Article Snippet: As an alternative to pre-training the drug encoder on matabolite properties and toxicity, we have studied the use of the 3D Infomax encoding for drugs.
Techniques: