Review Article
A Survey of Computational Intelligence Techniques in Protein Function Prediction
Table 6
Summary of computational intelligence techniques in prediction of nuclear/GPC receptor.
| | Reference | CI techniques | Prediction | Performance | Datasets |
| | [102] | SVM | NR | Overall accuracy: 82.6%–97.5% | Amino acid composition and dipeptide composition | | [103] | SVM | NR | Overall accuracy: 96% | 4-tuple residue composition | | [104] | SVM | NR | Overall accuracy: 99.6% | Pseudoamino acid composition | | [105] | SVM | NR | Accuracy: 98% | Pseudoamino acid composition | | [106] | SVM | NR | Accuracy: 97% | Amino acid composition, dipeptide composition, and physicochemical property | | [107] | Fuzzy -NN | NR | Overall accuracy: 93% | Pseudoamino acid composition with physicochemical and statistical features | | [108] | SVM | GPCR | Overall accuracy: 99.5% | Dipeptide composition of amino acids | | [21] | SVM | GPCR | Overall accuracy: 89.8%–96.4% | Amino acid composition and dipeptide composition | | [109] | SVM | GPCR | Overall accuracy: 99.6% | Pseudoamino acid composition | | [110] | Adaboost | GPCR | Overall accuracy: 96.4% and MCC: 0.930 | Pseudoamino acid composition with approximate entropy and hydrophobicity patterns | | [111] | PCA | GPCR | Overall accuracy: 80.47–99.5% | Sequence derived features |
|
|
