Prediction of dihedral angle regions in tertiary protein structures using neural networks

• 主要作者: Soh, Teng Chye.
• 其他作者: Tan Ching Wai
• 格式: Final Year Project
• 語言: English
• 出版: 2009
• 主題: DRNTU::Engineering::Computer science and engineering::Computer applications::Life and medical sciences
• 在線閱讀: http://hdl.handle.net/10356/19129

In bioinformatics, protein structure prediction is one of the most important goals and research. The three-dimensional structure of proteins is crucial as conformation plays an essential role in the wide ranging biological functions that they perform, hence by predicting and knowing the structure, we will be able to receive information on the function of the protein structure. In this project, we approach tertiary structure prediction by predicting the dihedral angle region, in the goal of constructing the three-dimensional structure without using complicated technique such as X-Ray Crystallography as they are time consuming and expensive. A short introduction to protein structure and protein structure prediction was given, followed by the methods used and discussion of the predicted results. Parameters extracted from Protein Data Bank (PDB) files were passed into a sequence alignment application ‘Lobster’, and then translated into dihedral angles. Various processing on these data with PSI-BLAST and PSIPRED were done, and fed into different neural networks.