EFFICIENT ALGORITHMS FOR PROTEIN TERTIARY STRUCTURE PREDICTION

这个子项目是众多研究子项目之一

This subproject is one of many research subprojects utilizing the resources provided by a Center grant funded by NIH/NCRR. The subproject and investigator (PI) may have received primary funding from another NIH source, and thus could be represented in other CRISP entries. The institution listed is for the Center, which is not necessarily the institution for the investigator. The structure of the molecules determines their possible reactions. Structural genomics studies protein structures and infers their functions based on structure. Protein threading is a comparative proteomic approach that determines an unknown proteins tertiary structure. However, it is difficult to cope with the inefficiency of this approach without compromising accuracy, especially for structure search in large protein databases. Currently, protein structure is predicted with ~80% accuracy. Other techniques such as X-ray crystallography and NMR spectroscopy are expensive and have low throughput. The goal of this proposed research in bioinformatics is to develop efficient parameterized algorithms for protein tertiary structure prediction. By identifying small parameters from the analysis of protein sequence and structure properties, parameterized approaches have the advantage of being very efficient, i.e., having low computational cost compared to the other traditional approaches such as approximation algorithms and statistical approaches. The parameterized approach could determine a large number of protein structures in a high throughput mode. The specific aims of the proposed research include the following: Aim 1: We design and implement efficient parameterized algorithms for protein tertiary structure prediction. Implementations will be made publicly available through a web services interface. Using sample techniques, from existing protein structure databases, the algorithms accuracy will be analyzed and compared to other available algorithms. Aim 2: Based on biological data provided by the mentor and other publicly-accessible sources, the parameterized algorithms will be improved to increase their accuracy with the goal of exceeding the current benchmark of an 80% predictive rate. Aim 3: Applying the implemented algorithms to the mentors data sets, we will predict protein tertiary structures which can be used to improve mutant protein stability in mutagenesis studies. The proposed research could provide useful information to tremendously reduce the time and expenses on doing biological experiments on blind prediction. Combined with physico-chemical analysis of protein structures, the proposed research has the potential to enable important biological discoveries, which could positively impact scientific discovery in the areas of biological science such as agricultural plant genetics, new pharmaceuticals design, and new protein production related to human health and disease.