Extreme Learning Machine for Predicting HLA-Peptide Binding

摘要: Machine learning techniques have been recognized as powerful tools for from data. One of the most popular techniques, Back-Propagation (BP) Artificial Neural Networks, can be used a computer model to predict peptides binding Human Leukocyte Antigens (HLA). The major advantage computational screening is that it reduces number wet-lab experiments need performed, significantly reducing cost and time. A recently developed method, Extreme Learning (ELM), which has superior properties over BP investigated accomplish such tasks. In our work, we found ELM good as, if not better than, in term time complexity, accuracy deviations across experiments, – importantly prevention over-fitting prediction peptide HLA.