Skip to main content

Advertisement

Table 2 Comparison of Nepal with other alignment methods

From: Derivative-free neural network for optimizing the scoring functions associated with dynamic programming of pairwise-profile alignment

  Remote [0,20)a (1405 files) Medium [20,40)a (1790 files) All [0,40)a (3195 files)
Sensitivity
 Nepal 0.5317 0.8343 0.7012
 Cosine 0.5045** 0.8246** 0.6838**
 CC 0.5135** 0.8269** 0.6891**
 MIQS 0.2775** 0.7316** 0.5319**
 BL62 0.2333** 0.6955** 0.4923**
Precision
 Nepal 0.5031 0.8102 0.6751
 Cosine 0.4753** 0.7999** 0.6571**
 CC 0.4858** 0.8032** 0.6636**
 MIQS 0.2654** 0.7134** 0.5164**
 BL62 0.2317** 0.6902** 0.4885**
  1. Cosine, CC, MIQS, and BL62, indicate profile comparison methods with cosine similarity and correlation coefficient and sequence comparison methods with MIQS and BLOSUM62
  2. ** P < 0.01, Wilcoxon signed rank test with Bonferroni correction
  3. aSequence identity (%) of each division