Use this URL to cite or link to this record in EThOS:
Title: Computational prediction of mitochondrial protein targeting
Author: Wills, Stephen
ISNI:       0000 0001 3569 8902
Awarding Body: University of Reading
Current Institution: University of Reading
Date of Award: 2005
Availability of Full Text:
Access from EThOS:
This research involves development of a novel tool called SANCAP (Statistical ANalysis & ClassificAtion of Proteins -!sancapl) to predict mitochondrial protein targeting based on protein amino acid sequence using statistical methods. This combines the ANOVA F-Test with a variant of the Covariant Discriminant Algorithm, itself an evolution of the Mahalanobis Least Distance Algorithm. Both methods are used to analyse varying N-terminal and C-terminal subsequence lengths alongside whole protein sequences, with soluble proteins analysed separate from membrane proteins. These , statistical methods are combined using a logical system which improves upon the results of any individual statis!ical result. {' SANCAP has proven successful as a mitochondrial targeting prediction tool with prediction success rates of 84% for soluble Human mitochondrial proteins. Similar results are generated when predicting extracellular proteins and chloroplast proteins. The prediction parameters defined for these locations and organisms can be used to analyse previously unseen protein sequences, and a website was created to provide fast userfriendly access to the SANCAP tool. In addition to predicting subcellular location of an individual protein or a list of proteins, the website also presents the results of predicting mitochondrial soluble and mitochondrial membrane proteins within the Human proteome. The final aspect of this research is that all prediction parameters and dataset generation is automated, as al,l program written during this research have been designed to run autonomously without compromising method efficacy.
Supervisor: Not available Sponsor: Not available
Qualification Name: Thesis (Ph.D.) Qualification Level: Doctoral
EThOS ID:  DOI: Not available