Table 1 Summary of 33 computational algorithms included in this study
From: Comprehensive assessment of computational algorithms in predicting cancer driver mutations
Classifier | Features | Method | Reference |
|---|---|---|---|
CADD | Conservation, epigenetic signals, functional predictions, genetic context, and published predictors | Linear kernel support vector machine | Rentzsch et al. [6] |
CanDrA | Structural, evolutionary, and genomic features, published predictors | Support vector machine | Mao et al. [7] |
CHASM | Structural, evolutionary, and genomic features | Random forest | Carter et al. [8] |
CTAT-cancer | TransFIC, fathmm, chasm, candra | Principal component analysis (PCA) | Bailey et al. [9] |
CTAT-population | SIFT, PolyPhen2, mutationAssessor, VEST | PCA | Bailey et al. [9] |
DANN | Conservation, epigenetic signals, functional predictions, and genetic context | Deep neural network | Quang et al. [10] |
DEOGEN2 | Evolutionary, protein, gene, pathway, PROVEAN | Random forest | Raimondi et al. [11] |
Eigen | Prediction scores of other tools, allele frequencies, epigenomic signals | Unsupervised spectral approach | Ionita-Laza et al. [12] |
Eigen-PC | Prediction scores of other tools, allele frequencies, epigenomic signals | Unsupervised spectral approach | Ionita-Laza et al. [12] |
FATHMM-disease | Sequence homology | Hidden Markov models | Shihab et al. [13] |
FATHMM-cancer | Sequence homology | Hidden Markov models | Shihab et al. [14] |
FATHMM-MKL | Conservation, epigenomic signals | Multiple kernel learning | Shihab et al. [15] |
FATHMM-XF | Conservation, genomic features, epigenomic signals | Multiple kernel learning | Rogers [16] |
GenoCanyon | Conservation, biochemical annotation | Posterior probability by unsupervised statistical learning | Lu et al. [17] |
Integrated_fitCons | Integrated epigenomic signals | INSIGHT | Gulko et al. [18] |
LRT | Sequence homology | Likelihood ratio test of codon neutrality | Chun et al. [19] |
M-CAP | Published predictors, conservation | Gradient boosting tree classifier | Jagadeesh et al. [20] |
MetaLR | Nine prediction scores and allele frequencies in 1000G | Logistic regression | Dong et al. [21] |
MetaSVM | Nine prediction scores and allele frequencies in 1000G | Radial kernel support vector machine | Dong et al. [21] |
MPC | Regional missense constraint, missense badness, polyphen2 | Logistic regression | Samocha et al. [22] |
MutationAssessor | Sequence homology | Combinatorial entropy formalism | Reva et al. [23] |
MutationTaster2 | Conservation, genetic context, regulatory features | Naïve Bayes classifier | Schwarz et al. [24] |
MutPred | Protein structural and functional properties, conservation, SIFT | Random forest | Li et al. [25] |
MVP | Sequence and structural features, published predictors, conservation | Deep neural network | Qian et al. [26] |
Polyphen2_HDIV | Eight sequence-based and three structure-based predictive features | Naïve Bayes classifier | Adzhubei et al. [27] |
Polyphen2_HVAR | Eight sequence-based and three structure-based predictive features | Naïve Bayes classifier | Adzhubei et al. [27] |
PrimateAI | Sequence homology | Deep residual neural network | Sundaram et al. [28] |
PROVEAN | Sequence homology | Delta alignment score | Choi et al. [29] |
REVEL | Published predictors | Random forest | Ioannidis et al. [30] |
SIFT | Sequence homology based on PSI-BLAST | Position-specific scoring matrix | Ng et al. [31] |
SIFT4G | Sequence homology based on Smith-Watermann | Position-specific scoring matrix | Vaser et al. [32] |
TransFIC | SIFT, Polyphen2, mutationAssessor | Transformed functional impact scores | Gonzalez-Perez [33] |
VEST4 | Amino acid-related features, DNA context, conservation, protein structure | Random forest | Carter et al. [34] |