Given a protein sequence and the C£\ coordinates on its backbone, the all-atom protein backbone reconstruction problem (PBRP) is to reconstruct the backbone by
its 3D coordinates of N, C and O atoms. In the past few decades, many methods have been proposed for solving PBRP, such as ab initio, homology modeling, SABBAC,
Wang¡¦s method, Chang¡¦s method, BBQ (Backbone Building from Quadrilaterals) and Chen¡¦s method. Chen found that, if they can choose the correct prediction tool
to build the 3D coordinates of the desired atoms, the RMSD may be improved. In this thesis, we propose a method for solving PBRP based on Chen¡¦s method. We
use tool preference classification on each atom of the residue, where the classification model is generated by SVM (Support Vector Machine). We rebuild the backbone by
combing the prediction results of all atoms in all residues. The data sets used in our experiments are CASP7, CASP8 and CASP9, which have 65, 52 and 63 proteins, respectively. These data sets contain nonstandard amino acids as well as standard ones. We improve the average RMSDs of Chen¡¦s results in some cases. The average
RMSDs of our method are 0.3496 in CASP7, 0.3084 in CASP8 and 0.3286 in CASP9.
| Identifer | oai:union.ndltd.org:NSYSU/oai:NSYSU:etd-0911112-132025 |
| Date | 11 September 2012 |
| Creators | Wu, Hsin-Fang |
| Contributors | Chia-Ning Yang, Kuo-Tsung Tseng, Chang-Biau Yang, Shih-Chung Chen, Yow-ling Shine |
| Publisher | NSYSU |
| Source Sets | NSYSU Electronic Thesis and Dissertation Archive |
| Language | English |
| Detected Language | English |
| Type | text |
| Format | application/pdf |
| Source | http://etd.lib.nsysu.edu.tw/ETD-db/ETD-search/view_etd?URN=etd-0911112-132025 |
| Rights | user_define, Copyright information available at source archive |
Page generated in 0.0023 seconds