Table 1.
Tool | Algorithm | Species | Encoding scheme | Evaluation strategy | URL/stand-alone package | Histone or nonhistone | Benchmark dataset | Option of batch prediction | Window size | Published year | Reference |
---|---|---|---|---|---|---|---|---|---|---|---|
Qiu et al. | SVM | Human Mouse |
PWAA | Jackknife test | No | Histone | 101 Histone proteins, 169 positive, 847 negative | No | 31 | 2017 | [13] |
CKSAAP CrotSite |
SVM | Human Mouse |
CKSAAP PseAAC |
Jackknife test | 123.206.31.171/ CKSAAP CrotSite/(not available) | Histone | 101 Histone proteins, 169 positive, 847 negative | No | 31 | 2017 | [15] |
iKcr-PseEns | Ensemble RF | Human Mouse |
PseAAC | Jackknife test | http://www.jci-bioinfo.cn/iKcr-PseEns (available) | Histone | 55 histones, 169 positive 46 Histones, 866 negative |
Yes | 35 | 2018 | [16] |
LightGBM-CroSite | LightGBM | Human Mouse |
BE PWAA EBGW kNN PsePSSM |
Jackknife test | https://github.com/QUST-AIBBDRC/LightGBM-CroSite/ | Histone | 101 Histone proteins, 159 positive, 847 negative | No | 31 | 2020 | [39] |
iCrotoK-PseAAC | ANN | Mixed | PseAAC RPRIM PRIM SM SVV FV AAPIV RAAPIV |
10-fold cross-validation | No | Mixed | 378 Positive, 500 negative | No | 41 | 2019 | [14] |
Deep-Kcr | CNN RF LB NB LR |
Human | CKSAAP PWAA AAindex CTD EBGW |
10-fold cross-validation | http://lin-group.cn/server/Deep-Kcr (available) | Mixed | 3734 Proteins, 9964 positive, 9964 negative | No | 31 | 2020 | [17] |
Abbreviations: EBGW, encoding based on grouped weight; kNN, k nearest neighbors, PsePSSM, pseudo-position specific scoring matrix; CTD, composition, transition and distribution; RPRIM, position relative incidence matrix; SVV, site vicinity vector; FV, frequency vector; AAPIV, accumulative absolute position incidence vector; RAAPIV, reverse accumulative absolute position incidence vector; NB, naïve Bayes; LB, logitboost.