Skip to main content

Thank you for visiting nature.com. You are using a browser version with limited support for CSS. To obtain the best experience, we recommend you use a more up to date browser (or turn off compatibility mode in Internet Explorer). In the meantime, to ensure continued support, we are displaying the site without styles and JavaScript.

Tools for drug discovery

In silico screening with chemical informatics

Imagine evaluating the efficacy and safety of new drugs without having to lift a pipette. Advances in in silico drug development are beginning to make this possible.

The SYBYL range of computational chemistry software from Tripos of St Louis, Missouri, will design compound libraries, select the best binding partners and screen potential drug molecules for safety and efficacy on the basis of known protein target and/or ligand structures — all in silico. A new advanced protein-modelling module predicts the structure of proteins of unknown structure using related proteins of known structure as the models. “It rapidly and accurately provides a valuable protein structure to be used for in silico docking experiments,” says Kathleen Mensler, vice-president of marketing and corporate development at Tripos. Another new product, Surflex-Sim, helps find new classes of compounds that are structurally similar to already known, suboptimal, lead compounds, but that may be more active or safer.

Modelling software from Cresset BioMolecular Discovery in Letchworth, UK, uses surface properties or 'molecular fields' of proteins and other molecules to predict intermolecular interactions. “We know that structurally diverse compounds can bind to the same protein site and elicit the same response, but this has never been fully explained by molecular structure,” says Sally Rose, Cresset's director of business development. “Our view is that molecular fields provide vital information for understanding and predicting activity,” she says. The fields are derived by calculating the interaction energy of a 'probe' atom carrying a positive, negative or neutral charge with the compound. “Our fields summarize the key binding information, and we locate the most important regions around a ligand where the fields are strongest and binding is expected,” says Rose. The original FieldScreen software for virtual screening has been joined by FieldTemplater, designed to find three-dimensional bioactive conformations, and FieldAlign, a molecular-alignment tool for comparing the three-dimensional structures of two molecules.

Line up: Cresset's FieldAlign program compares molecular structures. Credit: CRESSET BIOMOLECULAR DISCOVERY

A pioneer of in silico modelling, De Novo Pharmaceuticals in Cambridge, UK, recently announced a new focus for its drug-discovery programme — finding drug targets in various metabolic diseases. De Novo has upgraded SkelGen, its proprietary in silico drug-design platform, with the addition of Reflex, which takes into account the rotational flexibility of amino acids in the target's active site. This better evaluates the shape of the active site and its interaction with small-molecule ligands, so that the modelling more closely resembles the dynamics of real protein–ligand interactions.

C.S.

Rights and permissions

Reprints and Permissions

About this article

Cite this article

In silico screening with chemical informatics. Nature 446, 221 (2007). https://doi.org/10.1038/446221a

Download citation

Search

Quick links

Nature Briefing

Sign up for the Nature Briefing newsletter — what matters in science, free to your inbox daily.

Get the most important science stories of the day, free in your inbox. Sign up for Nature Briefing