Fig. 2: Performance of the VideoMol framework on multiple drug discovery tasks.
From: A molecular video-derived foundation model for scientific drug discovery
a The ROC (Receiver Operating Characteristic) curves of ImageMol and VideoMol on 10 main types of biochemical kinases with balanced scaffold split. The x-axis and y-axis represent FPR (False Positive Rate) and TPR (True Positive Rate), respectively. b The RMSE (Root Mean Squared Error) and MAE (Mean Absolute Error) performance of ImageMol and VideoMol on 10 GPCR with balanced scaffold split. c The ROC curves of ImageMol and VideoMol on 6 molecular property prediction benchmarks with scaffold split. d The RMSE (FreeSolv, ESOL, Lipo) and MAE (QM7, QM8, QM9) performance of ImageMol and VideoMol with scaffold split. For each of presentation, the values of FreesSolv and QM7 are scaled down by a factor of 2 and 100, respectively, and the values of QM8 and QM9 are scaled up by a factor of 50 and 100, respectively. e The ROC-AUC (Area Under the Receiver Operating Characteristic Curve) performance of REDIAL-2020, ImageMol, and VideoMol on 11 SARS-CoV-2 datasets. Source data are provided as a Source Data file.
