We report a computational approach that integrates structural bioinformatics, molecular modelling and systems biology to construct a drug-target network on a structural proteome-wide scale. The approach has been applied to the genome of Mycobacterium tuberculosis (M.tb), the causative agent of one of today's most widely spread infectious diseases. The resulting drug-target interaction network for all structurally characterized approved drugs bound to putative M.tb receptors, we refer to as the 'TB-drugome'. The TB-drugome reveals that approximately one-third of the drugs examined have the potential to be repositioned to treat tuberculosis and that many currently unexploited M.tb receptors may be chemically druggable and could serve as novel anti-tubercular targets. Furthermore, a detailed analysis of the TB-drugome has shed new light on the controversial issues surrounding drug-target networks [1]-[3]. Indeed, our results support the idea that drug-target networks are inherently modular, and further that any observed randomness is mainly caused by biased target coverage. The TB-drugome (http://funsite.sdsc.edu/drugome/TB) has the potential to be a valuable resource in the development of safe and efficient anti-tubercular drugs. More generally the methodology may be applied to other pathogens of interest with results improving as more of their structural proteomes are determined through the continued efforts of structural biology/genomics.

MOTIVATION: Due to the limitations in experimental methods for determining binary interactions and structure determination of protein complexes, the need exists for computational models to fill the increasing gap between genome sequence information and protein annotation. Here we describe a novel method that uses structural models to reduce a large number of in silico predictions to a high confidence subset that is amenable to experimental validation. RESULTS: A two-stage evaluation procedure was developed, first, a sequence-based method assessed the conservation of protein interface patches used in the original in silico prediction method, both in terms of position within the primary sequence, and in terms of sequence conservation. When applying the most stringent conditions it was found that 20.5% of the data set being assessed passed this test. Secondly, a high-throughput structure-based docking evaluation procedure assessed the soundness of three dimensional models produced for the putative interactions. Of the data set being assessed, 8264 interactions or over 70% could be modelled in this way, and 27% of these can be considered 'valid' by the applied criteria. In all, 6.9% of the interactions passed both the tests and can be considered to be a high confidence set of predicted interactions, several of which are described. AVAILABILITY: http://bioinformatics.leeds.ac.uk/~bmb4sjc. SUPPLEMENTARY INFORMATION: Supplementary data are available at Bioinformatics online.

MOTIVATION: There are two main areas of difficulty in homology modelling that are particularly important when sequence identity between target and template falls below 50%: sequence alignment and loop building. These problems become magnified with automatic modelling processes, as there is no human input to correct mistakes. As such we have benchmarked several stand-alone strategies that could be implemented in a workflow for automated high-throughput homology modelling. These include three new sequence-structure alignment programs: 3D-Coffee, Staccato and SAlign, plus five homology modelling programs and their respective loop building methods: Builder, Nest, Modeller, SegMod/ENCAD and Swiss-Model. The SABmark database provided 123 targets with at least five templates from the same SCOP family and sequence identities </=50%. RESULTS: When using Modeller as the common modelling program, 3D-Coffee outperforms Staccato and SAlign using both multiple templates and the best single template, and across the sequence identity range 20-50%. The mean model RMSD generated from 3D-Coffee using multiple templates is 15 and 28% (or using single templates, 3 and 13%) better than those generated by Staccato and Salign, respectively. 3D-Coffee gives equivalent modelling accuracy from multiple and single templates, but Staccato and SAlign are more successful with single templates, their quality deteriorating as additional lower sequence identity templates are added. Evaluating the different homology modelling programs, on average Modeller performs marginally better in overall modelling than the others tested. However, on average Nest produces the best loops with an 8% improvement by mean RMSD compared to the loops generated by Builder.

MOTIVATION: The large-scale comparison of protein-ligand binding sites is problematic, in that measures of structural similarity are difficult to quantify and are not easily understood in terms of statistical similarity that can ultimately be related to structure and function. We present a binding site matching score the Poisson Index (PI) based upon a well-defined statistical model. PI requires only the number of matching atoms between two sites and the size of the two sites-the same information used by the Tanimoto Index (TI), a comparable and widely used measure for molecular similarity. We apply PI and TI to a previously automatically extracted set of binding sites to determine the robustness and usefulness of both scores. RESULTS: We found that PI outperforms TI; moreover, site similarity is poorly defined for TI at values around the 99.5% confidence level for which PI is well defined. A difference map at this confidence level shows that PI gives much more meaningful information than TI. We show individual examples where TI fails to distinguish either a false or a true site paring in contrast to PI, which performs much better. TI cannot handle large or small sites very well, or the comparison of large and small sites, in contrast to PI that is shown to be much more robust. Despite the difficulty of determining a biological 'ground truth' for binding site similarity we conclude that PI is a suitable measure of binding site similarity and could form the basis for a binding site classification scheme comparable to existing protein domain classification schema.

MOTIVATION: Protein assemblies are currently poorly represented in structural databases and their structural elucidation is a key goal in biology. Here we analyse clefts in protein surfaces, likely to correspond to binding 'hot-spots', and rank them according to sequence conservation and simple measures of physical properties including hydrophobicity, desolvation, electrostatic and van der Waals potentials, to predict which are involved in binding in the native complex. RESULTS: The resulting differences between predicting binding-sites at protein-protein and protein-ligand interfaces are striking. There is a high level of prediction accuracy (< or =93%) for protein-ligand interactions, based on the following attributes: van der Waals potential, electrostatic potential, desolvation and surface conservation. Generally, the prediction accuracy for protein-protein interactions is lower, with the exception of enzymes. Our results show that the ease of cleft desolvation is strongly predictive of interfaces and strongly maintained across all classes of protein-binding interface.

MOTIVATION: Given that association and dissociation of protein molecules is crucial in most biological processes several in silico methods have been recently developed to predict protein-protein interactions. Structural evidence has shown that usually interacting pairs of close homologs (interologs) physically interact in the same way. Moreover, conservation of an interaction depends on the conservation of the interface between interacting partners. In this article we make use of both, structural similarities among domains of known interacting proteins found in the Database of Interacting Proteins (DIP) and conservation of pairs of sequence patches involved in protein-protein interfaces to predict putative protein interaction pairs. RESULTS: We have obtained a large amount of putative protein-protein interaction (approximately 130,000). The list is independent from other techniques both experimental and theoretical. We separated the list of predictions into three sets according to their relationship with known interacting proteins found in DIP. For each set, only a small fraction of the predicted protein pairs could be independently validated by cross checking with the Human Protein Reference Database (HPRD). The fraction of validated protein pairs was always larger than that expected by using random protein pairs. Furthermore, a correlation map of interacting protein pairs was calculated with respect to molecular function, as defined in the Gene Ontology database. It shows good consistency of the predicted interactions with data in the HPRD database. The intersection between the lists of interactions of other methods and ours produces a network of potentially high-confidence interactions.

MOTIVATION: Identifying the location of ligand binding sites on a protein is of fundamental importance for a range of applications including molecular docking, de novo drug design and structural identification and comparison of functional sites. Here, we describe a new method of ligand binding site prediction called Q-SiteFinder. It uses the interaction energy between the protein and a simple van der Waals probe to locate energetically favourable binding sites. Energetically favourable probe sites are clustered according to their spatial proximity and clusters are then ranked according to the sum of interaction energies for sites within each cluster. RESULTS: There is at least one successful prediction in the top three predicted sites in 90% of proteins tested when using Q-SiteFinder. This success rate is higher than that of a commonly used pocket detection algorithm (Pocket-Finder) which uses geometric criteria. Additionally, Q-SiteFinder is twice as effective as Pocket-Finder in generating predicted sites that map accurately onto ligand coordinates. It also generates predicted sites with the lowest average volumes of the methods examined in this study. Unlike pocket detection, the volumes of the predicted sites appear to show relatively low dependence on protein volume and are similar in volume to the ligands they contain. Restricting the size of the pocket is important for reducing the search space required for docking and de novo drug design or site comparison. The method can be applied in structural genomics studies where protein binding sites remain uncharacterized since the 86% success rate for unbound proteins appears to be only slightly lower than that of ligand-bound proteins. AVAILABILITY: Both Q-SiteFinder and Pocket-Finder have been made available online at http://www.bioinformatics.leeds.ac.uk/qsitefinder and http://www.bioinformatics.leeds.ac.uk/pocketfinder