Award details

Classification and electronic capture of enzyme reaction mechanisms through MACiE in CMLReact

Reference	BB/C51320X/1
Principal Investigator / Supervisor	Dr John Mitchell
Co-Investigators / Co-Supervisors	Professor Peter Murray-Rust
Institution	University of Cambridge
Department	Chemistry
Funding type	Research
Value (£)	57,698
Status	Completed
Type	Research Grant
Start date	01/06/2005
End date	31/08/2006
Duration	15 months

Abstract

In principle, much of the data contained within, or linked to by, MACiE could be used to classify enzymes and their reactions. This includes overall reactions, important in the EC scheme, individual steps, mechanistic features, residue roles, three dimensional active site structure and sequence. In this project, we will concentrate on using the steps of the catalysed chemical reactions, and their mechanisms, for classification. This follows ideas presented to the Nomenclature Committee of IUBMB (NC-IUBMB) and IUPAC-IUBMB Joint Commission on Biochemical Nomenclature (JCBN) by our group in May 2004. We envisage extracting relevant data from MACiE, to be stored as fingerprints ¿ relatively short string containing information about each MACiE entry. Importantly, these can be made context-dependent by changing the weighting of the different bits, which allows us to move beyond a one size fits all model of enzyme classification. The end user would then control which features of the finger print were used. This would provide the flexibility to provide different classifications for different problems, depending on the questions at hand. The fingerprints will provide a fine-tuneable means of quantifying similarities, primarily between reaction steps, which can be used for clustering. Such classifications will make apparent the similarities between steps in diverse mechanisms and will provide a knowledge base for application in areas like drug design. This knowledge will allow the transfer of chemical strategies between different therapeutic areas. Our work will advance the systematic understanding of the relationships between enzymes and thus inform studies of molecular evolution. MACiE is a forerunner in using CMLReact to represent chemical mechanism. Although applied here to enzymes, this work is more generally applicable to the storage and retrieval of mechanistic and structural information throughout chemistry and biochemistry. The technologies prototyped here will increase the value of chemical data by allowing easy storage, searching, comparison and visualisation of mechanistic information. Data entry consists of the capture of the detailed stepwise mechanism of enzymatic reactions from publications. The steps connect local minima along the reaction path. In MACiE, multistep enzyme reactions can be described by snapshots of a single XML-based hypermolecule of the atoms, bonds and electrons. For each step the snapshot represents the position of the hypermolecule in the reaction toplogy. Snapshots are easily and rapidly created in conventional chemical structure editors by making the precise edits to change bonds, positions and charges in the reaction steps. The snapshots will be combined to give a formal and consistent description of the complete reaction. This can then be animated in 2D and 3D movies to show the changes in coordinates, bonding, charges and mass balance during the reaction steps. We have generated 2D movies in SVG (Scalable Vector Graphics) and 2D reactions can be animated with Jmol. The project will involve work to improve and increasingly to automate this process.

Summary

unavailable

Committee	Closed Committee - Biomolecular Sciences (BMS)
Research Topics	X – not assigned to a current Research Topic
Research Priority	X – Research Priority information not available
Research Initiative	EDF (e-science Development Fund) (EDF) [2003-2005]
Funding Scheme	X – not Funded via a specific Funding Scheme

I accept the terms and conditions of use (opens in new window)

export PDF file

back to list new search