Progress toward virtual screening for drug side effects

AMDE: a novel attention-mechanism-based multidimensional feature encoder for drug-drug interaction prediction

The properties of the drug may be altered by the combination, which may cause unexpected drug-drug interactions (DDIs). Prediction of DDIs provides combination strategies of drugs for systematic and effective treatment. In most of deep learning-based methods for predicting DDI, encoded information about the drugs is insufficient in some extent, which limits the performances of DDIs prediction. In this work, we propose a novel attention-mechanism-based multidimensional feature encoder for DDIs prediction, namely attention-based multidimensional feature encoder (AMDE). Specifically, in AMDE, we encode drug features from multiple dimensions, including information from both Simplified Molecular-Input Line-Entry System sequence and atomic graph of the drug. Data experiments are conducted on DDI data set selected from Drugbank, involving a total of 34 282 DDI relationships with 17 141 positive DDI samples and 17 141 negative samples. Experimental results show that our AMDE performs better than some state-of-the-art baseline methods, including Random Forest, One-Dimension Convolutional Neural Networks, DeepDrug, Long Short-Term Memory, Seq2seq, Deepconv, DeepDDI, Graph Attention Networks and Knowledge Graph Neural Networks. In practice, we select a set of 150 drugs with 3723 DDIs, which are never appeared in training, validation and test sets. AMDE performs well in DDIs prediction task, with AUROC and AUPRC 0.981 and 0.975. As well, we use Torasemide (DB00214) as an example and predict the most likely drug to interact with it. The top 15 scores all have been reported with clear interactions in literatures.

Rapid Identification of Inhibitors and Prediction of Ligand Selectivity for Multiple Proteins: Application to Protein Kinases

Rapid identification of inhibitors for a family of proteins and prediction of ligand specificity are highly desirable for structure-based drug design. However, sequentially docking ligands into each protein target with conventional single-target docking methods is too computationally expensive to achieve these two goals, especially when the number of the targets is large. In this work, we use an efficient ensemble docking algorithm for simultaneous docking of ligands against multiple protein targets. We use protein kinases, a family of proteins that are highly important for many cellular processes and for rational drug design, as an example to demonstrate the feasibility of investigating ligand selectivity with this algorithm. Specifically, 14 human protein kinases were selected. First, native docking calculations were performed to test the ability of our energy scoring function to reproduce the experimentally determined structures of the ligand-protein kinase complexes. Next, cross-docking calculations were conducted using our ensemble docking algorithm to study ligand selectivity, based on the assumption that the native target of an inhibitor should have a more negative (i.e., favorable) energy score than the non-native targets. Staurosporine and Gleevec were studied as examples of nonselective and selective binding, respectively. Virtual ligand screening was also performed against five protein kinases that have at least seven known inhibitors. Our quantitative analysis of the results showed that the ensemble algorithm can be effective on screening for inhibitors and investigating their selectivities for multiple target proteins.

Druggability of Coronary Artery Disease Risk Loci

BACKGROUND

Genome-wide association studies have identified multiple loci associated with coronary artery disease and myocardial infarction, but only a few of these loci are current targets for on-market medications. To identify drugs suitable for repurposing and their targets, we created 2 unique pipelines integrating public data on 49 coronary artery disease/myocardial infarction-genome-wide association studies loci, drug-gene interactions, side effects, and chemical interactions.

METHODS

We first used publicly available genome-wide association studies results on all phenotypes to predict relevant side effects, identified drug-gene interactions, and prioritized candidates for repurposing among existing drugs. Second, we prioritized gene product targets by calculating a druggability score to estimate how accessible pockets of coronary artery disease/myocardial infarction-associated gene products are, then used again the genome-wide association studies results to predict side effects, excluded loci with widespread cross-tissue expression to avoid housekeeping and genes involved in vital processes and accordingly ranked the remaining gene products.

RESULTS

These pipelines ultimately led to 3 suggestions for drug repurposing: pentolinium, adenosine triphosphate, and riociguat (to target CHRNB4, ACSS2, and GUCY1A3, respectively); and 3 proteins for drug development: LMOD1 (leiomodin 1), HIP1 (huntingtin-interacting protein 1), and PPP2R3A (protein phosphatase 2, regulatory subunit b-double prime, α). Most current therapies for coronary artery disease/myocardial infarction treatment were also rediscovered.

CONCLUSIONS

Integration of genomic and pharmacological data may prove beneficial for drug repurposing and development, as evidence from our pipelines suggests.

Docking-based inverse virtual screening: methods, applications, and challenges

Identifying potential protein targets for a small-compound ligand query is crucial to the process of drug development. However, there are tens of thousands of proteins in human alone, and it is almost impossible to scan all the existing proteins for a query ligand using current experimental methods. Recently, a computational technology called docking-based inverse virtual screening (IVS) has attracted much attention. In docking-based IVS, a panel of proteins is screened by a molecular docking program to identify potential targets for a query ligand. Ever since the first paper describing a docking-based IVS program was published about a decade ago, the approach has been gradually improved and utilized for a variety of purposes in the field of drug discovery. In this article, the methods employed in docking-based IVS are reviewed in detail, including target databases, docking engines, and scoring function methodologies. Several web servers developed for non-expert users are also reviewed. Then, a number of applications are presented according to different research purposes, such as target identification, side effects/toxicity, drug repositioning, drug-target network development, and receptor design. The review concludes by discussing the challenges that docking-based IVS needs to overcome to become a robust tool for pharmaceutical engineering.

Comparative Modeling of Drug Target Proteins☆

In this perspective, we begin by describing the comparative protein structure modeling technique and the accuracy of the corresponding models. We then discuss the significant role that comparative prediction plays in drug discovery. We focus on virtual ligand screening against comparative models and illustrate the state-of-the-art by a number of specific examples.

Virtual target screening: validation using kinase inhibitors

Computational methods involving virtual screening could potentially be employed to discover new biomolecular targets for an individual molecule of interest (MOI). However, existing scoring functions may not accurately differentiate proteins to which the MOI binds from a larger set of macromolecules in a protein structural database. An MOI will most likely have varying degrees of predicted binding affinities to many protein targets. However, correctly interpreting a docking score as a hit for the MOI docked to any individual protein can be problematic. In our method, which we term "Virtual Target Screening (VTS)", a set of small drug-like molecules are docked against each structure in the protein library to produce benchmark statistics. This calibration provides a reference for each protein so that hits can be identified for an MOI. VTS can then be used as tool for: drug repositioning (repurposing), specificity and toxicity testing, identifying potential metabolites, probing protein structures for allosteric sites, and testing focused libraries (collection of MOIs with similar chemotypes) for selectivity. To validate our VTS method, twenty kinase inhibitors were docked to a collection of calibrated protein structures. Here, we report our results where VTS predicted protein kinases as hits in preference to other proteins in our database. Concurrently, a graphical interface for VTS was developed.

Structure-based systems biology for analyzing off-target binding

Here off-target binding implies the binding of a small molecule of therapeutic interest to a protein target other than the primary target for which it was intended. Increasingly such off-targeting appears to be the norm rather than the exception, rational drug design notwithstanding, and can lead to detrimental side-effects, or opportunities to reposition a therapeutic agent to treat a different condition. Not surprisingly, there is significant interest in determining a priori what off-targets exist on a proteome-wide scale. Beyond determining putative off-targets is the need to understand the impact of such binding on the complete biological system, with the ultimate goal of being able to predict the phenotypic outcome. While a very ambitious goal, some progress is being made.

Structure-Based Approaches to Target Fishing and Ligand Profiling

Chemogenomics is an emerging interdisciplinary field aiming at identifying all possible ligands of all possible targets. If one groups targets in columns and ligands in rows, chemogenomic approaches to drug discovery just fill the interaction matrix. Since experimental data do not suffice, several computational methods are currently actively developed to supplement time-consuming and costly experiments. They are either designed to fill rows and thus profile a ligand towards a heterogeneous set of targets (target profiling) or to fill columns and thus identify novel ligands for an existing target (standard virtual screening). At the interface of both strategies are now true chemogenomic computational methods filling well defined areas in the matrix. The present review will focus on (protein) structure-based approaches and illustrates major advances in this novel exciting field which is supposed to massively impact rational drug design in the next decade.

Analysis of pharmacology data and the prediction of adverse drug reactions and off-target effects from chemical structure

Preclinical Safety Pharmacology (PSP) attempts to anticipate adverse drug reactions (ADRs) during early phases of drug discovery by testing compounds in simple, in vitro binding assays (that is, preclinical profiling). The selection of PSP targets is based largely on circumstantial evidence of their contribution to known clinical ADRs, inferred from findings in clinical trials, animal experiments, and molecular studies going back more than forty years. In this work we explore PSP chemical space and its relevance for the prediction of adverse drug reactions. Firstly, in silico (computational) Bayesian models for 70 PSP-related targets were built, which are able to detect 93% of the ligands binding at IC(50) < or = 10 microM at an overall correct classification rate of about 94%. Secondly, employing the World Drug Index (WDI), a model for adverse drug reactions was built directly based on normalized side-effect annotations in the WDI, which does not require any underlying functional knowledge. This is, to our knowledge, the first attempt to predict adverse drug reactions across hundreds of categories from chemical structure alone. On average 90% of the adverse drug reactions observed with known, clinically used compounds were detected, an overall correct classification rate of 92%. Drugs withdrawn from the market (Rapacuronium, Suprofen) were tested in the model and their predicted ADRs align well with known ADRs. The analysis was repeated for acetylsalicylic acid and Benperidol which are still on the market. Importantly, features of the models are interpretable and back-projectable to chemical structure, raising the possibility of rationally engineering out adverse effects. By combining PSP and ADR models new hypotheses linking targets and adverse effects can be proposed and examples for the opioid mu and the muscarinic M2 receptors, as well as for cyclooxygenase-1 are presented. It is hoped that the generation of predictive models for adverse drug reactions is able to help support early SAR to accelerate drug discovery and decrease late stage attrition in drug discovery projects. In addition, models such as the ones presented here can be used for compound profiling in all development stages.

In silico elucidation of the molecular mechanism defining the adverse effect of selective estrogen receptor modulators

Early identification of adverse effect of preclinical and commercial drugs is crucial in developing highly efficient therapeutics, since unexpected adverse drug effects account for one-third of all drug failures in drug development. To correlate protein–drug interactions at the molecule level with their clinical outcomes at the organism level, we have developed an integrated approach to studying protein–ligand interactions on a structural proteome-wide scale by combining protein functional site similarity search, small molecule screening, and protein–ligand binding affinity profile analysis. By applying this methodology, we have elucidated a possible molecular mechanism for the previously observed, but molecularly uncharacterized, side effect of selective estrogen receptor modulators (SERMs). The side effect involves the inhibition of the Sacroplasmic Reticulum Ca2+ ion channel ATPase protein (SERCA) transmembrane domain. The prediction provides molecular insight into reducing the adverse effect of SERMs and is supported by clinical and in vitro observations. The strategy used in this case study is being applied to discover off-targets for other commercially available pharmaceuticals. The process can be included in a drug discovery pipeline in an effort to optimize drug leads and reduce unwanted side effects.

Early identification of the side effects of preclinical and commercial drugs is crucial in developing highly efficient therapeutics, as unexpected side effects account for one-third of all drug failures in drug development and lead to drugs being withdrawn from the market. Compared with the experimental identification of off-target proteins that cause side effects, computational approaches not only save time and costs by providing a candidate list of potential off-targets, but also provide insight into understanding the molecular mechanisms of protein–drug interactions. In this paper we describe an integrated approach to identifying similar drug binding pockets across protein families that have different global shapes. In a case study, we elucidate a possible molecular mechanism for the observed side effects of selective estrogen receptor modulators (SERMs), which are widely used to treat and prevent breast cancer and other diseases. The prediction provides molecular insight into reducing the side effects of SERMs and is supported by clinical and biochemical observations. The strategy used in this case study is being applied to discover off-targets for other commercially available pharmaceuticals and to repurpose existing safe pharmaceuticals to treat different diseases. The process can be included in a drug discovery pipeline in an effort to optimize drug leads, reduce unwanted side effects, and accelerate development of new drugs.

GlamDock: Development and Validation of a New Docking Tool on Several Thousand Protein−Ligand Complexes

In this study, we present GlamDock, a new docking tool for flexible ligand docking. GlamDock (version 1.0) is based on a simple Monte Carlo with minimization procedure. The main features of the method are the energy function, which is a continuously differentiable empirical potential, and the definition of the search space, which combines internal coordinates for the conformation of the ligand, with a mapping-based description of the rigid body translation and rotation. First, we validate GlamDock on a standard benchmark, a set of 100 protein-ligand complexes, which allows comparative evaluation to existing docking tools. The results on this benchmark show that GlamDock is at least comparable in efficiency and accuracy to the best existing docking tools. The main focus of this work is the validation on the scPDB database of protein-ligand complexes. The size of this data set allows a thorough analysis of the dependencies of docking accuracy on features of the protein-ligand system. In particular, it allows a two-dimensional analysis of the results, which identifies a number of interesting dependencies that are generally lost or even misinterpreted in the one-dimensional approach. The overall result that GlamDock correctly predicts the complex structure in practically half of the cases in the scPDB is important not only for screening ligands against a particular protein but even more so for inverse screening, that is, the identification of the correct targets for a particular ligand.

Comparative Modeling of Drug Target Proteins

In this perspective, we begin by describing the comparative protein structure modeling technique and the accuracy of the corresponding models. We then discuss the significant role that comparative prediction plays in drug discovery. We focus on virtual ligand screening against comparative models and illustrate the state of the art by a number of specific examples.

kinDOCK: a tool for comparative docking of protein kinase ligands

KinDOCK is a new web server for the analysis of ATP-binding sites of protein kinases. This characterization is based on the docking of ligands already co-crystallized with other protein kinases. A structural library of protein kinase–ligand complexes has been extracted from the Protein Data Bank (PDB). This library can provide both potential ligands and their putative binding orientation for a given protein kinase. After protein–protein structural superposition, the ligands are transferred from the template complexes to the target protein kinase. The resulting complexes are evaluated using the program SCORE to compute a theoretical affinity. They can be dynamically visualized to allow a rapid mapping of important steric clashes and potential substitutions relevant for specificity and affinity. These characteristics allow a quick characterization of protein kinase active sites including conformation changes potentially required to accommodate particular ligands. Additionally, promising pharmacophores can be identified in the focussed library. These features will help to rationalize or optimize virtual screening (VS) on larger chemical compound libraries. The server and its documentation are freely available at .

Multiple ligand-binding modes in bacterial R67 dihydrofolate reductase

R67 dihydrofolate reductase (DHFR), a bacterial plasmid-encoded enzyme associated with resistance to the drug trimethoprim, shows neither sequence nor structural homology with the chromosomal DHFR. It presents a highly symmetrical toroidal structure, where four identical monomers contribute to the unique central active-site pore. Two reactants (dihydrofolate, DHF), two cofactors (NADPH) or one of each (R67*DHF*NADPH) can be found simultaneously within the active site, the last one being the reactive ternary complex. As the positioning of the ligands has proven elusive to empirical determination, we addressed the problem from a theoretical perspective. Several potential structures of the ternary complex were generated using the docking programs AutoDock and FlexX. The variability among the final poses, many of which conformed to experimental data, prompted us to perform a comparative scoring analysis and molecular dynamics simulations to assess the stability of the complexes. Analysis of ligand-ligand and ligand-protein interactions along the 4 ns trajectories of eight different structures allowed us to identify important inter-ligand contacts and key protein residues. Our results, combined with published empirical data, clearly suggest that multipe binding modes of the ligands are possible within R67 DHFR. While the pterin ring of DHF and the nicotinamide ring of NADPH assume a stacked endo-conformation at the centre of the pore, probably assisted by V66, Q67 and I68, the tails of the molecules extend towards opposite ends of the cavity, adopting multiple configurations in a solvent rich-environment where hydrogen-bond interactions with K32 and Y69 may play important roles.

Reengineering the pharmaceutical industry by crash-testing molecules

The recent decline in drug approvals and the increase in late-stage failures indicate that the ability to generate and screen large numbers of molecules has not improved the drug pipeline. Perhaps the pharmaceutical industry should follow the example of the automotive industry and agree upon a shared modeling language with vendors and academics to enable integration of predictive computational tools across the industry. This will then enable the virtual 'crash-testing' of drugs before synthesis, biological testing and, most importantly, clinical trials. This represents an ambitiously progressive approach using the models for simulating every stage of the drug discovery and development process. Combining the relevant computational algorithms into a grand unified model would enable prioritization of the best ideas before pursuing a discovery program, selecting a target or synthesizing a molecule. The successful application of these virtual crash-testing principles by any of its current proponents could revitalize the pharmaceutical industry so that failure is avoided.

Structural basis for the GSK-3beta binding affinity and selectivity against CDK-2 of 1-(4-aminofurazan-3yl)-5-dialkylaminomethyl-1H-[1,2,3] triazole-4-carboxylic acid derivatives

A novel structural class of glycogen synthase kinase-3beta inhibitors is modeled using quantum mechanics, automated docking, and molecular dynamics simulations. The proposed binding modes identify important hydrogen bonds and salt-bridges with the ATP-binding pocket of the kinase. The modeled complexes justify the observed structure-activity relationships and provide a structural basis for the high selectivity of these inhibitors against cyclin dependent kinase-2.

Rapid computational identification of the targets of protein kinase inhibitors

We describe a method for rapidly computing the relative affinities of an inhibitor for all individual members of a family of homologous receptors. The approach, implemented in a new program, SCR, models inhibitor-receptor interactions in full atomic detail with an empirical energy function and includes an explicit account of flexibility in homology-modeled receptors through sampling of libraries of side chain rotamers. SCR's general utility was demonstrated by application to seven different protein kinase inhibitors: for each inhibitor, relative binding affinities with panels of approximately 20 protein kinases were computed and compared with experimental data. For five of the inhibitors (SB203580, purvalanol B, imatinib, H89, and hymenialdisine), SCR provided excellent reproduction of the experimental trends and, importantly, was capable of identifying the targets of inhibitors even when they belonged to different kinase families. The method's performance in a predictive setting was demonstrated by performing separate training and testing applications, and its key assumptions were tested by comparison with a number of alternative approaches employing the ligand-docking program AutoDock (Morris et al. J. Comput. Chem. 1998, 19, 1639-1662). These comparison tests included using AutoDock in nondocking and docking modes and performing energy minimizations of inhibitor-kinase complexes with the molecular mechanics code GROMACS (Berendsen et al. Comput. Phys. Commun. 1995, 91, 43-56). It was found that a surprisingly important aspect of SCR's approach is its assumption that the inhibitor be modeled in the same orientation for each kinase: although this assumption is in some respects unrealistic, calculations that used apparently more realistic approaches produced clearly inferior results. Finally, as a large-scale application of the method, SB203580, purvalanol B, and imatinib were screened against an almost full complement of 493 human protein kinases using SCR in order to identify potential new targets; the predicted targets of SB203580 were compared with those identified in recent proteomics-based experiments. These kinome-wide screens, performed within a day on a small cluster of PCs, indicate that explicit computation of inhibitor-receptor binding affinities has the potential to promote rapid discovery of new therapeutic targets for existing inhibitors.

Binding MOAD (Mother Of All Databases)

Binding MOAD (Mother of All Databases) is the largest collection of high-quality, protein-ligand complexes available from the Protein Data Bank. At this time, Binding MOAD contains 5331 protein-ligand complexes comprised of 1780 unique protein families and 2630 unique ligands. We have searched the crystallography papers for all 5000+ structures and compiled binding data for 1375 (26%) of the protein-ligand complexes. The binding-affinity data ranges 13 orders of magnitude. This is the largest collection of binding data reported to date in the literature. We have also addressed the issue of redundancy in the data. To create a nonredundant dataset, one protein from each of the 1780 protein families was chosen as a representative. Representatives were chosen by tightest binding, best resolution, etc. For the 1780 "best" complexes that comprise the nonredundant version of Binding MOAD, 475 (27%) have binding data. This significant collection of protein-ligand complexes will be very useful in elucidating the biophysical patterns of molecular recognition and enzymatic regulation. The complexes with binding-affinity data will help in the development of improved scoring functions and structure-based drug discovery techniques. The dataset can be accessed at http://www.BindingMOAD.org.

LEA3D: a computer-aided ligand design for structure-based drug design

We present an improved version of the program LEA developed to design organic molecules. Rational drug design involves finding solutions to large combinatorial problems for which an exhaustive search is impractical. Genetic algorithms provide a tool for the investigation of such problems. New software, called LEA3D, is now able to conceive organic molecules by combining 3D fragments. Fragments were extracted from both biological compounds and known drugs. A fitness function guides the search process in optimizing the molecules toward an optimal value of the properties. The fitness function is build up by combining several independent property evaluations, including the score provided by the FlexX docking program. One application in de novo drug design is described. The example makes use of the structure of Mycobacterium tuberculosis thymidine monophosphate kinase to generate analogues of one of its natural substrates. Among 22 tested compounds, 17 show inhibitory activity in the micromolar range.

Inhibition studies with rationally designed inhibitors of the human low molecular weight protein tyrosine phosphatase

The human low molecular weight protein tyrosine phosphatase (HCPTP) is ubiquitously expressed as two isoforms in a wide range of human cells and may be involved in regulating the metastatic nature of epithelial tumors. A homology model is presented for the HCPTP-B isoform based on known X-ray crystal structures of other low molecular weight PTPs. A comparison of the two isoform structures indicates the possibility of developing isoform-specific inhibitors of HCPTP. Molecular dynamics simulations with CHARMM have been used to study the binding modes of the known adenine effector and phosphate in the active site of both isoforms. This analysis led to the design of the initial lead compound, based on an azaindole ring moiety, which was then also evaluated computationally. A comparison of these simulations indicates the need for a phosphonate group on the indole and provides insight into inhibitor binding modes. Compounds with varying degrees of structural similarity to the azaindole have been synthesized and tested for inhibition with each isoform. These molecular systems were examined with the program AutoDock, and comparisons made with the kinetics and the explicit simulations to validate AutoDock as a screening tool for potential inhibitors. Two compounds were experimentally found to have sub-millimolar inhibition, but the greater solubility of one reinforces the need for experimental testing alongside computational analysis.

Comparing Ligand Interactions with Multiple Receptors via Serial Docking

Standard uses of ligand-receptor docking typically focus on the association of candidate ligands with a single targeted receptor, but actual applications increasingly require comparisons across multiple receptors. This study demonstrates that comparative docking to multiple receptors can help to select homology models for virtual compound screening and to discover ligands that bind to one set of receptors but not to another, potentially similar, set. A serial docking algorithm is furthermore described that reduces the computational costs of such calculations by testing compounds against a series of receptor structures and discarding a compound as soon as it fails to satisfy specified bind/no bind criteria for each receptor. The algorithm also realizes substantial efficiencies by taking advantage of the fact that a ligand typically binds in similar conformations to similar receptors. Thus, once detailed docking has been used to fit a ligand into the first of a series of similar receptors, much less extensive calculations can be used for the remaining structures.

Electrostatic Potential of Nucleotide-free Protein is Sufficient for Discrimination Between Adenine and Guanine-specific Binding Sites

Despite sharing many common features, adenine-binding and guanine-binding sites in proteins often show a clear preference for the cognate over the non-cognate ligand. We have analyzed electrostatic potential (ESP) patterns at adenine and guanine-binding sites of a large number of non-redundant proteins where each binding site was first annotated as adenine/guanine-specific or non-specific from a survey of primary literature. We show that more than 90% of ESP variance at the binding sites is accounted for by only two principal component ESP vectors, each aligned to molecular dipoles of adenine and guanine. Projected on these principal component vectors, the adenine/guanine-specific and non-specific binding sites, including adenine-containing dinucleotides, show non-overlapping distributions. Adenine or guanine specificities of the binding sites also show high correlation with the corresponding electrostatic replacement (cognate by non-cognate ligand) energies. High correlation coefficients (0.94 for 35 adenine-binding sites and 1.0 for 20 guanine-binding sites) were obtained when adenine/guanine specificities were predicted using the replacement energies. Our results demonstrate that ligand-free protein ESP is an excellent indicator for discrimination between adenine and guanine-specific binding sites and that ESP of ligand-free protein can be used as a tool to annotate known and putative purine-binding sites in proteins as adenine or guanine-specific.

Ligand selectivity and competition between enzymes in silico

In a cell, there are many possibilities for cross interactions between enzymes and small molecules, arising from the similarities in the structures of the metabolites and the flexibility in binding of protein active sites. Despite this promiscuity, the cognate partners must be able to recognize each other in vivo, for the cell to function efficiently. This study examines the basis of this selectivity in recognition using standard docking calculations and finds significant improvement when proteins and ligands are cross-docked. We find that cognate molecules rarely form the most stable complexes and that specificity may be driven either by recognition of the substrate by the enzyme or the recognition of the enzyme by the substrate. Despite limitations of the in silico methods, especially the scoring functions, these calculations highlight the need to consider cross reactions in the cell and suggest that localization and compartmentalization must be important factors in the evolution of complex cells. However, the inherent promiscuity of these interactions can also benefit an organism, by facilitating the evolution of new functions from old ones. The results also suggest that high-throughput screening should involve not just a panel of small molecules, but also a panel of proteins to test for cross-reactivity.

Predicting undesirable drug interactions with promiscuous proteins in silico

Although computational tools have been used to predict toxic responses resulting from molecules binding either as substrates or inhibitors to proteins, there are complications to be resolved. Some proteins appear promiscuous in their ability to bind a diverse array of hydrophobic molecules. This promiscuity arises from the binding site simultaneously accommodating more than one molecule, multiple separate binding sites, protein flexibility, or a combination of all these properties. With the availability of more crystal structures for these non-target proteins, we should be able to predict binding in silico with a greater accuracy, thus avoiding or managing toxic side effects, therefore ultimately improving the success of drug discovery.

What is the relevance of bioinformatics to pharmacology?

Although bioinformatics achieved prominence because of its central role in genome data storage, management and analysis, its focus has shifted as the life sciences exploit these data. In pharmacology, genomic, transcriptomic and proteomic data are being used in the quest for drugs that fulfill unmet medical needs, are disease modifying or curative and are more effective and safer than current drugs. Bioinformatics is used in drug target identification and validation and in the development of biomarkers and toxicogenomic and pharmacogenomic tools to maximize the therapeutic benefit of drugs. Now that the 'parts list' of cellular signalling pathways is available, integrated computational and experimental programmes are being developed, with the goal of enabling in silico pharmacology by linking the genome, transcriptome and proteome to cellular pathophysiology.

A Comprehensive Docking Study on the Selectivity of Binding of Aromatic Compounds to Proteins

Generally, computer-aided drug design is focused on screening of ligand molecules for a single protein target. The screening of several proteins for a ligand is a relatively new application of molecular docking. In the present study, complexes from the Brookhaven Protein Databank were used to investigate a docking approach of protein screening. Automated molecular docking calculations were applied to reproduce 44 protein-aromatic ligand complexes (31 different proteins and 39 different ligand molecules) of the databank. All ligands were docked to all different protein targets in altogether 12090 docking runs. Based on the results of the extensive docking simulations, two relative measures, the molecular interaction fingerprint (MIF) and the molecular affinity fingerprint (MAF), were introduced to describe the selectivity of aromatic ligands to different proteins. MIF and MAF patterns are in agreement with fragment and similarity considerations. Limitations and future extension of our approach are discussed.