Skip to Main Content
Understanding protein structures is vital to determining the function of a protein and its interaction with DNA, RNA and enzyme. The information about its conformation can provide essential information for drug design and protein engineering. While there are over a million known protein sequences, only a limited number of protein structures are experimentally determined. Hence, prediction of protein structures from protein sequences using computer programs is an important step to unveil proteins' three dimensional conformation and functions. As a result, prediction of protein structures has profound theoretical and practical influence over biological study. The explanation of how a decision is made during prediction is also important for improving protein structure prediction and guiding the "wet experiments". In this talk, we will show how to use machine learning methods to improve the accuracy of protein structure prediction and to interpret prediction results. We will report our research on using neural networks, support vector machines combined with decision tree and association rule for protein structure prediction, rule extraction and prediction interpretation. Evaluation and comparisons of various prediction and rule extraction systems will be presented and future research direction in this area will also be identified.