BMC Bioinformatics

Table 2 Performance comparison between the present method (NMF), LSTM, LA-kernel, and SW-PSSM

From: Application of nonnegative matrix factorization to improve profile-profile alignment features for fold recognition and remote homolog detection

Methods	Fold level		Superfamily level
	Mean ROC	Mean ROC₅₀	Mean ROC	Mean ROC₅₀
NMF	0.84	0.44	0.96	0.86
LSTM	0.70	0.25	0.77	0.39
LA-kernel	0.80	0.30	0.88	0.59
SW-PSSM(2.0, 10, 0.0)	0.85	0.43	0.96	0.83
SW-PSSM(3.0,0.75,1.5)	0.88	0.46	0.96	0.85

Back to article page

ISSN: 1471-2105

Contact us

General enquiries: journalsubmissions@springernature.com