Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

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

Figure 2

ROC 50 scores of various methods for fold recognition at the fold level. The x-axis represents the ROC50 score and the y-axis represents the proportion of proteins with better performance than the corresponding ROC50 score. The mean ROC50 score using NMF features is 0.44, while the corresponding figures are 0.21, 0.20, and 0.10 for original PPA features, HHsearch, and PSI-BLAST, respectively. At > 0.95 ROC50 scores, 22% of protiens with NMF features correctly detect remotely related proteins, whereas original PPA features detect almost none.

Back to article page

BMC Bioinformatics

ISSN: 1471-2105

Contact us