Table 9 The comparison of the PRMxAI with previous predictors on the same data set used by the previous predictors [39]
Year | Author | Prediction method | Algorithm | ACC (%) | SP (%) | SEN (%) | MCC |
|---|---|---|---|---|---|---|---|
2006 | Chen et al. [30] | MeMo | SVM | 68 | 99 | 38 | 0.46 |
2009 | Shao et al. [33] | BPB-PPMS | SVM | 56 | 100 | 12 | 0.25 |
2009 | Shien et al. [31] | MASA | SVM | 65 | 99 | 31 | 0.41 |
2012 | Shi et al. [42] | PMeS | SVM | 58 | 73 | 43 | 0.16 |
2014 | Qiu et al. [32] | iMethyl-PseAAC | SVM | 59 | 100 | 18 | 0.30 |
2016 | Wen et al. [43] | PSSMe | SVM | 72 | 83 | 60 | 0.44 |
2017 | Wei et al. [35] | MePred-RF | RF | 69 | 97 | 41 | 0.46 |
2020 | Chaudhari et al. [38] | DeepRMethylSite | CNN, LSTM | 79.42 | 84.47 | 75.08 | 0.60 |
2021 | Lumbanraja et al. [39] | SSMFN | CNN, LSTM | 81.15 | 82.40 | 80.00 | 0.62 |
2023 | Proposed model | PRMxAI | RF | 83.84 | 87.61 | 80.76 | 0.68 |