3DICE: interpretable 3D cross-modal learning for drug-target interaction prediction and large-scale drug discovery.

As shown in Table S6 , using an alternative conformer seed produces no statistically significant change in hard-prediction metrics (ACC: t = −0.884, P = .427; MCC: t = −1.19, P = .301), but a highly significant decrease in AUROC ( t = −23.9, P = .0000181), with an overall prediction flip rate of 5.45%.

also in 132,142 other papers