Dissertation > Excellent graduate degree dissertation topics show

Prediction of β-turns in Proteins

Author: QianGang
Tutor: YuanZheMing
School: Hunan Agricultural University
Course: Bioinformatics
Keywords: β-turn prediction homology information PDB NetTurnP BTMapping
CLC: Q51
Type: Master's thesis
Year: 2012
Downloads: 28
Quote: 0
Read: Download Dissertation

Abstract


β-turn is a secondary protein structure type which actually belongs to the coil region as one of the three basic secondary structure elements in proteins, and it is important in protein folding, protein stability and molecular recognition processes. To date, various methods have been put forward to predict β-turns, they can be divided into statistical methods and machine learning techniques. Normally, machine learning methods have better performances. Interestingly, none of them have tried directly to map the structures of pre-existing homologues from structural databases like RCSB PDB to the protein to be predicted. Given the large size of PDB (>70.000structures), it is actually of high possibility to find a structural homologue for a newly identified sequence.In this work, we present a new method that predicts β-turns by combining homology information extracted from PDB with the results predicted by NetTurnP that is a de novo β-turn predictor using two layers of neural networks. Two datasets, the golden set BT426and the self-constructed dataset EVA937, are used to assess our method. For each sequence in both datasets, only homologues deposited earlier than the sequence in PDB are employed. We have achieved Matthews correlation coefficients (MCCs) of0.56,0.52respectively, which are higher than those obtained by NetTurnP alone of0.50,0.46, and the prediction accuracies (Qtotal) obtained using our method are81.4%and80.4%separately, while NetTurnP alone achieves78.2%and77.3%. The results confirm that combining the homology information with state-of-the-art β-turn predictors like NetTurnP can significantly improve the prediction accuracy.In order to evaluate the performance of our method under various conditions when the BLAST hits of the query sequences have different sequence identity levels, we vary Imax from0.2to1.0with a step of0.1to control the selection of the BLAST hits that are more than100residues in length. The results shows that by employing the homology information from the BLAST hits, the accuracy of NetTurnP has been improved even at low Imax. The parameter ’byDate’ in our program controls the selection of BLAST hits, only the hits deposited earlier than the date represented by "byDate" are adopted for structure mapping. Therefore, set with different’byDate’values, we try to test if our method will have different prediction accuracies. Finally we get the conclusion that as the ’byDate’ value increases, MCC and Qtotal both increase, which means the prediction accuracy becomes higher. We believe the reason for this is that as the ’byDate’value gets nearer, more homologues to a query sequence can be found through BLAST, and it is of higher possibility to find optimal homologues for structure mapping to increase the prediction accuracy. A Java program called BTMapping has been written to implement our method, which is freely available at http://www.bio530.weebly.com together with the related datasets.

Related Dissertations

  1. Original Mechanism Research on the Antihyperglycaemic and Antihyperlipidaetic Activity of the Potentilla Discolor Bunge,R285
  2. Discolor herbology and Pharmacodynamics Study,R285
  3. PDB and medicinal plants are among ISSR analysis and HPLC fingerprint analysis,S567.239
  4. Study on Hypoglycemic Effect of Potentilla Discolor Bunge and Effect on Metabolism of Glucose in Liver Cell,R285.5
  5. PDB on diabetic rats and its mechanisms,R285
  6. Experimental study of the hypoglycemic effect and its mechanism of PDB,R285
  7. Cross-platform Map Displaying in Embedded Environment,P208
  8. Research on Sequence Similarity Search Using Fractal Features of Proteins,TP391.41
  9. Effect of the Sencond Domain Deletion on the Interaction between Yelp3 and Elp4 and Other Proteins,Q51
  10. Comparison of the structural model and its application in the prediction of transcription factor binding sites in the protein-DNA,Q51
  11. Prediction of Protein-protein Interaction and DNA Methylation Sites Based on Increment of Diversity Combined with Quadratic Discriminant Analysis,Q51
  12. Regeneration and Preparation of IMAC Proteinchip,Q51
  13. The Comparison of Anti-angiogenic Activities among the Lj-RGD3 Toxin Protein and Its’ Five Forms of Mutation Proteins,Q51
  14. Phylogenetic Analysis of SERPINA Subfamily and SERPINA3n Expression in Microtus Fortis,Q51
  15. Study on the Mass Transfer Process of Mambrane Crystallization of Protein,Q51
  16. The Study on Modeling Protein Side-chain with HMM and Its Application,Q51
  17. Research on the Structural Transformation of Regenerated Silk Fibroin,Q51
  18. Study on Law and Predicting Methods of Protein-protein Interaction,Q51
  19. Predicting Protein-protein Interaction and Study on hub Protein Classification and Interaction Law,Q51
  20. Conjunction of Spider Silk Protein by Trans-Splicing of Split-Intein,Q51
  21. Kinetic Analysis of Trans-Splicing of Novel Split-Inteins in Vitro,Q51

CLC: > Biological Sciences > Biochemistry > Protein
© 2012 www.DissertationTopic.Net  Mobile