Use this URL to cite or link to this record in EThOS:
Title: Integrated bioinformatic approaches to the prediction of protein-protein interactions
Author: Lee, Lisa Li Chun
Awarding Body: University of Manchester
Current Institution: University of Manchester
Date of Award: 2011
Availability of Full Text:
Access from EThOS:
Access from Institution:
Protein-protein interactions (PPIs) play a fundamental role in many biological processes such as signal transduction from the extracelluar space to cytosol. Functions of less characterized proteins can often be deduced from PPI networks. Various sequence-based approaches were taken to predicting and understanding potential PPIs using bioinformatic means. Initially, the mirror tree method was comprehensively examined to derive a robust approach for PPI predictions. The analysis has revealed that mirror tree is extremely sensitive to many factors especially sequence diversity and the selection of orthologues. Indeed, higher sequence diversity improves the predictive power of the approach. In an attempt to improve prediction accuracy, various speciation signal correction methods were evaluated and the RNA-based approaches appear to be more effective in removing the speciation signal and ultimately produce more accurate predictions. The utility of mirror tree was further extended for domain-domain interactions in fibrillin-1. However, due to the low sequence diversity of the orthologues, poor prediction results were obtained. Furthermore, a residue based method utilizing the mutual information (MI) statistic was evaluated for intramolecular protein interaction predictions. Similar to the mirror tree method, removal of the background signal occurring from common ancestry improves the prediction accuracy. When MI of a third position was incorporated to facilitate the interaction prediction between two contacting positions, the prediction quality was increased. Moreover, in order to identify clusters consisting of three contacting residues, position combinations with the highest significant partial correlation coefficients were extracted and their atomic distances were compared to assess the accuracy of the prediction. Lastly, an analysis was carried out to study the association between PRINTS fingerprints and functionally important interaction sites in seven G protein-coupled receptor families. More than 50% of the functional sites acquired from literature were found to be in close proximity to fingerprint motifs. In the surface patch analysis, over 80% of the functional sites were shown to overlap a motif cluster. Overall, the approaches taken in this thesis have tackled interaction predictions from various directions and keenly provide some insights for protein-protein interactions and evolution.
Supervisor: Hubbard, Simon ; Attwood, Teresa Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available
Keywords: protein-protein interaction ; mutual information ; GPCR