Table 5. Open in new tab Performance of...

Table 5.

Performance of the benchmark method, PocketMiner, and P2Rank was evaluated across different subsets of the CryptoBench test set.

Method	Dataset	AUC	AUPRC	ACC	FPR	TPR	MCC	F1 Score
pLM-NN	CB-full	0.86	0.36	0.93	0.05	0.48	0.39	0.92
pLM-NN	CB-PM	0.88	0.43	0.93	0.04	0.52	0.44	0.93
PocketMiner	CB-PM	0.76	0.19	0.82	0.16	0.51	0.22	0.78
pLM-NN	CB-P2RANK-apo	0.88	0.42	0.93	0.04	0.51	0.43	0.93
P2RANK	CB-P2RANK-apo	0.81	0.21	0.85	0.14	0.62	0.27	0.81
P2RANK	CB-P2RANK-holo	0.89	0.34	0.85	0.15	0.84	0.38	0.81

Method	Dataset	AUC	AUPRC	ACC	FPR	TPR	MCC	F1 Score
pLM-NN	CB-full	0.86	0.36	0.93	0.05	0.48	0.39	0.92
pLM-NN	CB-PM	0.88	0.43	0.93	0.04	0.52	0.44	0.93
PocketMiner	CB-PM	0.76	0.19	0.82	0.16	0.51	0.22	0.78
pLM-NN	CB-P2RANK-apo	0.88	0.42	0.93	0.04	0.51	0.43	0.93
P2RANK	CB-P2RANK-apo	0.81	0.21	0.85	0.14	0.62	0.27	0.81
P2RANK	CB-P2RANK-holo	0.89	0.34	0.85	0.15	0.84	0.38	0.81

In the first evaluation round, the benchmark method was assessed using the full CryptoBench test set (CB-full). In the second round, both the benchmark method and PocketMiner were evaluated on a subset that included only structures for which PocketMiner did not fail (CB-PM); see Supplementary Material for details. In the third round, the benchmark method and P2rank were tested on a subset consisting solely of single-chain apo structures (CB-P2RANK-apo). Lastly, P2Rank was evaluated on holo structures (CB-P2RANK-holo), which are the counterparts of the apo structures from the CB-P2RANK-apo subset. The comparison between P2Rank’s performance on CB-P2RANK-apo and CB-P2RANK-holo highlights the performance drop when identifying CBSs using a method not specialized for detecting such sites. The F1 score was computed using a weighted average.

This Feature Is Available To Subscribers Only