pLMSNOSite: an ensemble-based approach for predicting protein S-nitrosylation sites by integrating supervised word embedding and embedding from pre-trained protein language model

Table 7 Performance comparison of pLMSNOSite against other existing approaches using the independent test set

Predictors	TP	FP	TN	FN	ACC	SN	SP	MCC	AUROC
GPS-SNO	99	825	2337	253	0.693	0.281	0.739	0.014	0.523
iSNO-PseAAC	101	768	2394	251	0.710	0.287	0.757	0.031	–
SNOSite	235	1749	1413	117	0.469	0.668	0.447	0.069	–
DeepNitro	202	776	2386	148	0.737	0.578	0.737	0.222	0.731
PreSNO	211	733	2431	141	0.752	0.604	0.769	0.252	0.756
pLMSNOSite	258	718	2446	93	0.769	0.735	0.773	0.340	0.754

The highest values in each column are highlighted in bold
Note that the values for other approaches were adopted from PreSNO. Although same independent test set was used for all the approaches, there is a slight variation in the number of total positive and negative sites. Nevertheless, the integrity of comparison is not compromised at all

ISSN: 1471-2105