Identifying Anticancer Peptides from Amino Acid Sequences: A Novel Approach Based on Deep Neural Networks

Cancer, also known as malignant tumors, originates primarily from cellular mutations causing uncontrolled growth. These abnormal cells can metastasize to other body parts, leading to organ failure. Anticancer peptides are innovative drugs that target and disrupt cancer cells, with fewer side effects than chemotherapy and targeted therapy. Despite previous development of anticancer peptide prediction tools, their predictive performance needs improvement. To enhance accuracy and applicability and reduce development costs, this study introduces the Deep Neural Networks Anticancer Peptide Prediction Model (DAPPM). Experimental results show that the proposed model achieves high accuracy on the test dataset, with 92.2% accuracy, 91.5% sensitivity, 92.9% specificity, and 84.4% Matthews correlation coefficient. The study demonstrates superior predictive capabilities and method feasibility compared to previous research efforts.