シェフィールド大学のタンパク質計算ソフトウェア

GOLD is a program for calculating the docking modes of small molecules in protein binding sites and is provided as part of the GOLD Suite, a package of programs for structure visualisation and manipulation (Hermes), for protein-ligand docking (GOLD) and for post-processing (GoldMine) and visualisation of docking results. Hermes acts as a hub for many of CCDC's products, for more information please refer to the Hermes product page.

The product of a collaboration between the University of Sheffield, GlaxoSmithKline plc and CCDC, GOLD is very highly regarded within the molecular modelling community for its accuracy and reliability.

GOLD features include:

• A genetic algorithm (GA) for protein-ligand docking
• An easy to use interface with interactive docking set-up via Hermes
• A comprehensive docking set-up wizard
• Full ligand flexibility
• Partial protein flexibility, including protein side chain and backbone flexibility for up to ten user-defined residues
• Energy functions partly based on conformational and non-bonded contact information from the CSD
• A variety of constraint options
• Improved flexible ring handling
• Automatic consideration of cavity bound water molecules
• Improved handling and control of metal coordination geometries
• Improved parameterisation for kinases and heme-containing proteins
• Automatic derivation of GA settings for particular ligands
• A choice of GoldScore, ChemScore, Astex Statistical Potential (ASP) or Piecewise Linear Potential (PLP) scoring functions
• Extensive options for customising or implementing new scoring functions through a Scoring Function Application Programming Interface, allowing users to modify the GOLD scoring-function mechanism in order to either: implement their own scoring function or enhance existing scoring functions; customise docking output
• A ChemScore Receptor Depth Scaling (RDS) rescore option so that the score attributed to hydrogen bonds is scaled depending on the depth in the binding pocket
• Automatic rescoring with an alternate scoring function at the end of a docking run.

GOLD has been fully validated against 305 diverse and extensively checked protein-ligand complexes from the PDB (CCDC/Astex Test Set). 72% of GOLD's top-ranked solutions were found to be accurate using stringent success criteria. A further 85 diverse, high quality drug-like complexes have been validated; GOLD reproduces the observed binding mode within 2.0 Angstroms for 81% of the structures (Astex Diverse Set). More recently the Astex Diverse Set has been used to analyse GOLD's cross-docking performance (the Astex Non Native Set).

GOLD's genetic algorithm parameters are optimised for virtual screening applications. GOLD is optimised for parallel execution on processor networks; a distributed version of GOLD is available for use on commercial PC GRID systems