Computational Chemistry
The chemical laws which determine the routes and rates of metabolism of foreign compounds are poorly understood and cannot easily be predicted in vivo.
Approaches to model metabolism and relate molecular properties to metabolic outcomes promises to greater help guide drug design and toxicity prediction.
Quantitative Structure-Activity Relationship (QSAR) analysis is a well-established tool for optimising a series of bio-active molecules. The technique is statistically based and aimed at extracting the maximum information from biological test data (in this case metabolic data) on compounds of known structure and physicochemical properties. If the physicochemical properties can be calculated or predicted then the relationships identified in the analysis can be used predictively.
QSAR analysis is greatly aided by computational chemistry, which provides a widely applicable source of calculable properties and a detailed description of a wide range of molecular environments. The data matrices generated by computational chemistry are best interrogated by multivariate PR techniques that can be used to classify biological activity parameters in terms of chemical properties of the molecule using dimension-reduction techniques.
Computational facilities are available in Biomolecular Medicine, including access to a computational cluster and a high-performance computing.
We have now adopted an approach whereby the "metabolic fate" of compounds (determined principally by novel NMR methods) is related, by use of pattern recognition and other multivariate statistical methods, to their calculated molecular physicochemical properties. This combined NMR-pattern recognition and computational chemistry approach appears to allow useful predictive models of drug metabolism to be built for the first time and the extraction of the chemical rules that determine metabolic fate in vivo.
By gaining an understanding of the molecular physicochemical interactions that are important for site-recognition on drug metabolising enzymes it should be possible to improve the design of novel therapeutic agents and increase both their efficacy and reduce their toxicity.
Key Recent Publications
Warne MA, Nicholson JK, Lindon JC, Guiney PD, Gartland KP. 2009. A QSAR investigation of dermal and respiratory chemical sensitizers based on computational chemistry properties. SAR QSAR Environ Res 20(5-6):429-51.
