Originally a BBSRC-EPSRC funded project, joint with Leeds University,
and now funded by the University of Glasgow.
This project is developing the TOPS system for visualization, database searching, comparison and machine learning of protein structural topology. The project has created a database of protein protein topology, enhanced with sequence and function-related information. Fast and efficient algorithms have been developed for structure comparison, motif discovery and pattern matching, and versions are being developed to include sequence and function information. In addition visualization facilities for TOPS diagrams have been developed. The most speculative part of the project, is the investigation of machine learning for structure-sequence-function relationships, which aims to learn patterns linking topology to sequence, and topology and sequence to function. Interesting results will be valuable to efforts to predict protein structure and function from sequence, and these problems remain key challenges of direct relevance to projects in structural and functional genomics.
Glasgow: David Gilbert (PI) and Gilleain Torrance
Leeds: David Westhead (PI) and Ioannis Michalopoulos
TOPS project website (database and software systems, plus TOPS resources: www.tops.leeds.ac.uk
Software systems at the Bioinformatics Research Centre, Glasgow: