Files and links (1)
PDFOpen Access (License Unspecified), Open Access
Dissertation
Prediction of protein structures and interaction sites
Doctor of Philosophy (Ph.D.), Drexel University
2001
DOI:
https://doi.org/10.17918/00007863
Abstract
This thesis project was focused on homology-based prediction of protein structures from amino acid sequences, and prediction of potential protein-protein interface residues. In the project, a threading method based on dynamical programming was developed and tested, which is integrated in COBLATH, a program our group developed for fold recognition and query-template alignment. Results of COBLATH show that the combination of two existing methods, PSI-BLAST and threading, leads to significant enhancement in the success rate of fold recognition and alignment accuracy. The thesis project was involved in structural annotation on Mycoplasma genitaanium and Saccharomyces cerevisiae genomes by COBLATH. The results of the annotations are presented. Our group selected a nonhomologous set of 744 complex-forming protein pairs from PDB and identified the interface residues of these proteins. Based on this protein set split for training and testing, a neural-network-based method (PPISP) for prediction of protein-protein interface residues in a given structure was developed in the thesis project. This method takes a surface residue's solvent accessibility and the position specific sequence profile of itself and its 19 nearest surface neighbors as input for its prediction. A prediction accuracy of 70% is obtained when the method is tested on 129 pairs of nonhomologous complex-forming proteins. The development of COBLATH and PPISP opens the possibility that protein-protein interaction sites may be predicted on a genome-wide scale. To that end, we have built structural models for yeast protein-protein complexes based on the predicted structures of the individual proteins by COBLATH and the experimental complex structures from other species. Structures of these complexes, FTase, GGTase, and Ras-Gap complexes in particular, are analyzed to gain understanding of their biological specificities. To make progress toward establishing the physical basis of protein stability, we searched for parallels between residue-residue potential of mean force (PMF) in proteins and in liquids. We found some remarkable similarities.
Metrics
14 File views/ downloads
22 Record Views
Details
- Title
- Prediction of protein structures and interaction sites
- Creators
- Yibing Shan
- Contributors
- Huan Xiang Zhou (Advisor) - Drexel University, Drexel University (1970-)
- Awarding Institution
- Drexel University
- Degree Awarded
- Doctor of Philosophy (Ph.D.)
- Publisher
- Drexel University; Philadelphia, Pennsylvania
- Number of pages
- xi, 177 pages
- Resource Type
- Dissertation
- Language
- English
- Academic Unit
- College of Arts and Sciences; Drexel University
- Other Identifier
- 991021889068204721