Search

There is currently great interest in comparing protein-ligand docking programs. A review of recent comparisons shows that it is difficult to draw conclusions of general applicability. Statistical hypothesis testing is required to ensure that differences in pose-prediction success rates and enrichment rates are significant. Numerical measures such as root-mean-square deviation need careful interpretation and may profitably be supplemented by interaction-based measures and visual inspection of dockings. Test sets must be of appropriate diversity and of good experimental reliability. The effects of crystal-packing interactions may be important. The method used for generating starting ligand geometries and positions may have an appreciable effect on docking results. For fair comparison, programs must be given search problems of equal complexity (e.g. binding-site regions of the same size) and approximately equal time in which to solve them. Comparisons based on rescoring require local optimization of the ligand in the space of the new objective function. Re-implementations of published scoring functions may give significantly different results from the originals. Ostensibly minor details in methodology may have a profound influence on headline success rates.

We present a large test set of protein–ligand complexes for the purpose of validating algorithms that rely on the prediction of protein–ligand interactions. The set consists of 305 complexes with protonation states assigned by manual inspection. The following checks have been carried out to identify unsuitable entries in this set: (1) assessing the involvement of crystallographically related protein units in ligand binding; (2) identification of bad clashes between protein side chains and ligand; and (3) assessment of structural errors, and/or inconsistency of ligand placement with crystal structure electron density. In addition, the set has been pruned to assure diversity in terms of protein–ligand structures, and subsets are supplied for different protein-structure resolution ranges. A classification of the set by protein type is available. As an illustration, validation results are shown for GOLD and SuperStar. GOLD is a program that performs flexible protein–ligand docking, and SuperStar is used for the prediction of favorable interaction sites in proteins. The new CCDC/Astex test set is freely available to the scientific community (http://www.ccdc.cam.ac.uk). Proteins 2002;49:457–471. © 2002 Wiley-Liss, Inc.

Recently, the development of computer programs which permit the de novo design of molecular structures satisfying a set of steric and chemical constraints has become a burgeoning area of research and many operational systems have been reported in the literature. Experience with PRO-LIGAND-the de novo design methodology embodied in our in-house molecular design and simulation system PRO-METHEUS-has suggested that the addition of a genetic algorithm (GA) structure refinement procedure can 'add value' to an already useful tool. Starting with the set of designed molecules as an initial population, the GA can combine features from both high- and low-scoring structures and, over a number of generations, produce individuals of better score than any of the starting structures. This paper describes how we have implemented such a procedure and demonstrates its efficacy in improving two sets of molecules generated by different de novo design projects.

Optimal sequence threading can be used to recognize members of a library of protein folds which are closely related in 3-D structure to the native fold of an input test sequence, even when the test sequence is not significantly homologous to the sequence of any member of the fold library. The methods provide an alignment between the residues of the test sequence and the residue positions in a template fold. This alignment optimizes a score function, and the predicted fold is the highest scoring member of the library of folds. Most score functions contain a pairwise interaction energy term. This, coupled with the need to introduce gaps into the alignment, means that the optimization problem is NP hard. We report a comparison between two heuristic optimization algorithms used in the literature, double dynamic programming and an iterative algorithm based on the so-called frozen approximation. These are compared in terms of both the ranking of likely folds and the quality of the alignment produced.

We implemented a novel approach to score water mediation and displacement in the protein-ligand docking program GOLD. The method allows water molecules to switch on and off and to rotate around their three principal axes. A constant penalty, sigma(p), representing the loss of rigid-body entropy, is added for water molecules that are switched on, hence rewarding water displacement. We tested the methodology in an extensive validation study. First, sigma(p) is optimized against a training set of 58 protein-ligand complexes. For this training set, our algorithm correctly predicts water mediation/displacement in approximately 92% of the cases. We observed small improvements in the quality of the predicted binding modes for water-mediated complexes. In the second part of this work, an entirely independent set of 225 complexes is used. For this test set, our algorithm correctly predicts water mediation/displacement in approximately 93% of the cases. Improvements in binding mode quality were observed for individual water-mediated complexes.

The cytochromes P450 (P450s) are a family of heme-containing monooxygenase enzymes involved in a variety of functions, including the metabolism of endogenous and exogenous substances in the human body. During lead optimization, and in drug development, many potential drug candidates are rejected because of the affinity they display for drug-metabolising P450s. Recently, crystal structures of human enzymes involved in drug metabolism have been determined, significantly augmenting the prospect of using structure-based design to modulate the binding and metabolizing properties of compounds against P450 proteins. An important step in the application of structure-based metabolic optimization is the accurate prediction of docking modes in heme binding proteins. In this paper we assess the performance of the docking program GOLD at predicting the binding mode of 45 heme-containing complexes. We achieved success rates of 64% and 57% for Chemscore and Goldscore respectively; these success rates are significantly lower than the value of 79% observed with both scoring functions for the full GOLD validation set. Re-parameterization of metal-acceptor interactions and lipophilicity of planar nitrogen atoms in the scoring functions resulted in a significant increase in the percentage of successful dockings against the heme binding proteins (Chemscore 73%, Goldscore 65%). The modified scoring functions will be useful in docking applications on P450 enzymes and other heme binding proteins.

The Chemscore function was implemented as a scoring function for the protein-ligand docking program GOLD, and its performance compared to the original Goldscore function and two consensus docking protocols, "Goldscore-CS" and "Chemscore-GS," in terms of docking accuracy, prediction of binding affinities, and speed. In the "Goldscore-CS" protocol, dockings produced with the Goldscore function are scored and ranked with the Chemscore function; in the "Chemscore-GS" protocol, dockings produced with the Chemscore function are scored and ranked with the Goldscore function. Comparisons were made for a "clean" set of 224 protein-ligand complexes, and for two subsets of this set, one for which the ligands are "drug-like," the other for which they are "fragment-like." For "drug-like" and "fragment-like" ligands, the docking accuracies obtained with Chemscore and Goldscore functions are similar. For larger ligands, Goldscore gives superior results. Docking with the Chemscore function is up to three times faster than docking with the Goldscore function. Both combined docking protocols give significant improvements in docking accuracy over the use of the Goldscore or Chemscore function alone. "Goldscore-CS" gives success rates of up to 81% (top-ranked GOLD solution within 2.0 A of the experimental binding mode) for the "clean list," but at the cost of long search times. For most virtual screening applications, "Chemscore-GS" seems optimal; search settings that give docking speeds of around 0.25-1.3 min/compound have success rates of about 78% for "drug-like" compounds and 85% for "fragment-like" compounds. In terms of producing binding energy estimates, the Goldscore function appears to perform better than the Chemscore function and the two consensus protocols, particularly for faster search settings. Even at docking speeds of around 1-2 min/compound, the Goldscore function predicts binding energies with a standard deviation of approximately 10.5 kJ/mol.

Many proteins undergo small side chain or even backbone movements on binding of different ligands into the same protein structure. This is known as induced fit and is potentially problematic for virtual screening of databases against protein targets. In this report we investigate the limits of the rigid protein approximation used by the docking program, GOLD, through cross-docking using protein structures of influenza neuraminidase. Neuraminidase is known to exhibit small but significant induced fit effects on ligand binding. Some neuraminidase crystal structures caused concern due to the bound ligand conformation and GOLD performed poorly on these complexes. A 'clean' set, which contained unique, unambiguous complexes, was defined. For this set, the lowest energy structure was correctly docked (i.e. RMSD1.5 A away from the crystal reference structure) in 84% of proteins, and the most promiscuous protein (1mwe) was able to dock all 15 ligands accurately including those that normally required an induced fit movement. This is considerably better than the 70% success rate seen with GOLD against general validation sets. Inclusion of specific water molecules involved in water-mediated hydrogen bonds did not significantly improve the docking performance for ligands that formed water-mediated contacts but it did prevent docking of ligands that displaced these waters. Our data supports the use of a single protein structure for virtual screening with GOLD in some applications involving induced fit effects, although care must be taken to identify the protein structure that performs best against a wide variety of ligands. The performance of GOLD was significantly better than the GOLD implementation of ChemScore and the reasons for this are discussed. Overall, GOLD has shown itself to be an extremely good, robust docking program for this system.