BMC Bioinformatics

Table 1 The evaluation of single feature performance on the training set

From: An improved DNA-binding hot spot residues prediction method by exploring interfacial neighbor properties

Features	SEN	SPE	PRE	F1	MCC	ACC	AUC
All features	0.726	0.841	0.763	0.744	0.571	0.793	0.839
no INP1-sASA	0.661	0.761	0.661	0.661	0.423	0.720	0.777
no Psi	0.548	0.841	0.708	0.618	0.411	0.720	0.780
no INP2-CsRSA	0.645	0.761	0.656	0.650	0.408	0.713	0.796
no Eig	0.645	0.784	0.678	0.661	0.433	0.727	0.809
no CtASA	0.661	0.795	0.695	0.678	0.460	0.740	0.807
no HbCN	0.742	0.795	0.719	0.730	0.535	0.773	0.826
no DN	0.694	0.830	0.741	0.717	0.529	0.773	0.811

The lowest value in each column is shown in italics

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com