Q. H. Nguyen, T-H Nguyen-Vo, Trang T. T. Do, Susanto Rahardja, and B. P. Nguyen∗
Short-length AMPs have been demonstrated to have intensified antimicrobial activities toward wider spectrums of microbes. Therefore, exploration of novel and promising short AMPs is highly essential in developing various types of antimicrobial drugs or treatments. Besides, experimental approaches, several computational approaches have been developed to improve screening efficiency. Although these existing computational methods achieved satisfactory performance, there is a large room for model improvement. In this study, we proposed iAMP-DL, a prediction framework for identifying short AMPs. The model was constructed using long short-term memory incorporated with convolutional neural networks. To fairly assess the model performance, we compare our models with the existing state-of-the-art methods.
The comparative analysis's results confirmed iAMP-DL's performance outperformed all existing state-of-the-art methods. Furthermore, the experiments were repeated ten times to observe the variation in the prediction efficiency of our proposed method. The results demonstrated that the iAMP-DL is an effective, robust, and stable framework that can be used to detect promising short-AMP.