Peptide-based drug predictions for cancer therapy using deep learning

Background: Therapeutic drugs used in cancer treatment have ineffectiveness and resistance to drug action problems. Anticancer peptides (ACPs) are selective and toxic to cancer cells and quickly produced. Thus, ACPs can be a satisfactory substitute for therapeutic drugs. We developed AI4ACP, a user-friendly web-server ACP predictor that can predict the anticancer property of query peptides, thus promoting the discovery of peptides with anticancer activity. Result: Our results revealed that the performance of our ACP predictor trained using the new ACP collection was superior to that of the available high-performance ACP predictors. Conclusions: AI4ACP is a user-friendly web-server ACP predictor that can be used to determine whether a query sequence is an ACP. This tool can be beneficial for drug development for cancer treatment. AI4ACP is freely accessible at https://axp.iis.sinica.edu.tw/AI4ACP/ ### Competing Interest Statement The authors have declared no competing interest.