Milburn, Duncan; (2000) Computational Analysis of Functional Sites in Proteins. Doctoral thesis (Ph.D), UCL (University College London).

Text Computational_analysis_of_the_.pdf Download (18MB)

Abstract

Enzymes are the basis of life, fulfilling a wide spectrum of functional roles in all cells. Primary sequences of many enzymes are publicly available, and for some of these their three dimensional structure is known. This thesis presents a computational approach to help identify relatives of an enzyme sequence by enhancing conventional analysis techniques with the addition of structural information. The Web-based tool. Sequences Annotated by Structure (SAS), uses the FASTA algorithm to find potential relatives with known structures whose sequences are then coloured according to a range of physical and chemical properties such as secondary structure, active site and molecular interactions. On inspection, the results can be used to assign homologous relationships between sequences and allow functional inferences to be made, thus making the jump from sequence → structure → function. Two families of enzymes both utilising aspartic acids in their catalysis have been studied in more detail - the polymerase family and the aspartic protease family. In each case, SAS has been used alongside many structural tools including SSAP and TESS, shedding new light onto the relationships between family members. The polymerases are a diverse family of enzymes whose function is to replicate or repair sections of DNA or RNA. Despite little similarity between their sequences, detailed structural analysis led to the identification of a conserved region (called the 'palm') found in members of all four functional classes. The investigation into the sequences and structures of the aspartic protease family of enzymes showed that despite great functional diversity, there is some structural similarity between all observed members. In both studies, a relationship has been found between known structures and an industrially relevant protein sequence. A method for constructing active site templates from structures was applied to both families. The active site of the polymerase family was located in the 'palm' region of the enzyme and while essentially the same for all examples, different templates were necessary to represent unrelated members of the family efficiently using both TESS and SPASM programs. The active site of the aspartic protease family, situated between two domains in the structure, is much more rigid and a single TESS template was found to represent the site very well. However, both active site motifs were found to be highly sensitive and specific when compared to all of the structures in the Protein Data Bank. This work contributes to a library of three-dimensional active site templates currently being developed at UCL.

Download activity - last month

Export as JSONXMLCSV

Download activity - last 12 months

Export as JSONCSVXML

Downloads by country - last 12 months

Export as XMLCSVJSON

Archive Staff Only

View Item