|
| Method | Training set | Result/Output | Accuracy |
|
| 14 ML algorithms including RF, CART, SVM, K-nearest neighbor, Logistic Regression (LOGREG), Adaptive Boosting (AB) [157] | HemoPI-1, HemoPI-2 and HemoPI-3 | Predicts hemolytic nature with more accuracy than hemolytic activity | HemoPI-1: 94-95%
HemoP2/3- 75-77% |
| AI4AMP [158]: PC6 encoded Deep Learning Model | APD3, LAMP, CAMP3, DRAMP | Considers physicochemical properties of aminoacids as features | 90% |
| Deep-AmPEP30 [159]: CNN, Reduced AAC (RAAC) | Subset of a training set used in AmPEP [160] (experimentally validated AMPs from APD3, LAMP, CAMPR3) | Predicts AMP activity of short peptides from genomic DNA. | 77.13% |
| AntiBP [161]: ANN, Quantitative Matrices (QM), SVM | 486 peptides shorter than 61 residues from the APD database | Preference for certain residues at the N- and C-terminals that demarcate them from non antibacterial peptides | ANN based: 88.17%, QM based: 90.37% SVM based: 92.11% |
| AMPA [162]: Web application to assess the antimicrobial domain in protein. Can serve as template for AMP design | High throughput screening results from amino acid substitutions on bactenecin 2A | Quick discovery of new AMP patterns in protein sequences | accuracy of 85% and a sensitivity of 90% [163] |
| AVPpred [164]: SVM | 1245 experimentally validated antiviral peptides | Predicts highly effective AVPs | 86% |
| iAMP-2L [165]: Pseudo-amino acid composition, Fuzzy K-Nearest neighbor | Curated set of AMP and non-AMP peptides collected from Uniprot | Predicts if a given sequence is an AMP or not and if yes, the next classification is done on the type of AMP | 92.23% |
| AMP_Scanner [166]: Deep Neural Network with the Keras framework (http://www.keras.io) using a sequential model and a TensorFlow deep learning library back-end | APD vr.3 database | This model automatically extracts expert-free features and hence removes the reliance on domain experts for feature construction | 91.01% |
|
