A comprehensive study of small non-frameshift insertions/deletions in proteins and prediction of their phenotypic effects by a machine learning method (KD4i)

Table 6 Comparison of prediction performance for KD4i with the DDIG-in and SIFT-Indel methods

Algorithm	Accuracy	Sensitivity	Specificity	Precision	NPV	MCC
DDIG-in	83	-	-	-	-	0.67
SIFT-Indel	82	81	82	82	-	0.63
KD4i (SVM)	84	80	87	88	79	0.67
KD4i (ILP) (average)	79	89	69	75	85	0.59
KD4i (ILP) (final)	83	93	74	78	91	0.68

For the KD4i (ILP) implementation, the average performances in the 10-fold cross-validation and the performance of the final selected rule set (corresponding to fold 1 in Table 5) are provided. For DDIG-in and SIFT-Indel, the performances for combined insertions and deletions (as originally reported by the authors, for similar balanced data sets) are shown.

ISSN: 1471-2105