A Software Toolkit for de Novo Ligand Design

ICAMS, School of Chemistry, University of Leeds, Leeds, UK

Abstract

A new Software Toolkit, based on the SPROUT program, is under development. It is an interactive system that assists the chemists in several steps of the structure-based rational drug design process. The system is modularised and offers automatic methods for solving a number of problems in drug design. However, the user maintains control and is able to guide each module and modify its decisions as required. Six modules of the Toolkit are described below.
 
 

• The module CAnGAROO detects potential binding pockets of protein structures by detecting large inward-facing regions in the solvent accessible surface of the protein.
• HIPPO identifies favourable hydrogen bonding and hydrophobic regions within a binding pocket. The hydrogen bonding sites are directional and are used to define target sites for the position of potential ligand atoms.
• The module EleFAnT will select functional groups and position them at the target sites to form starting fragments for structure generation.
• SPIDeR generates skeletons that satisfy the steric constraints of a binding pocket by growing spacer fragments onto the start fragments and then connecting the resulting part skeletons together.
• MARABOU substitutes hetero atoms into the skeletons to generate molecules with the electrostatic properties that are complementary to those of the receptor site.
• Finally, ALLigaTOR will cluster and score the solutions to provide the user with an efficient tool for evaluating and navigating through the results.

The capability of the system for generating thought provoking ideas is demonstrated by examples.

Introduction

SPROUT is a computer program for constrained structure generation. It can generate molecules that can fit the steric and chemical constraints of a pharmacophore hypothesis or a specified protein receptor site. In the latter case the steric constraints consist of a volume derived from the active site of a protein together with some interaction sites within the volume. These interaction sites are called target sites and they are used to model different interaction types, e.g. hydrogen bonding.

The previously published version [1] of the program requires the definition of the target sites prior to running SPROUT. The target sites were treated as spheres of fixed radius around points that are specified in an input file. The new version uses a more sophisticated representation of the different types of target sites. Structure generation is now supported by additional modules which assist other stages of the molecule design process.

For a detailed description of each module, please follow the links in the abstract text above.

References

1. V. Gillet, A.P. Johnson, P. Mata, S. Sike, P. Williams, J. Comput.-Aided Mol. Design, 7 (1993) 127.

All correspondence should be addressed to Dr. A.P. Johnson

Return to the ECCC homepage.