Karina S. Machado

Mining flexible-receptor docking experiments to select promising protein receptor snapshots

BackgroundMolecular docking simulation is the Rational Drug Design (RDD) step that investigates the affinity between protein receptors and ligands. Typically, molecular docking algorithms consider receptors as rigid bodies. Receptors are, however, intrinsically flexible in the cellular environment. The use of a time series of receptor conformations is an approach to explore its flexibility in molecular docking computer simulations, but it is extensively time-consuming. Hence, selection of the most promising conformations can accelerate docking experiments and, consequently, the RDD efforts.ResultsWe previously docked four ligands (NADH, TCL, PIF and ETH) to 3,100 conformations of the InhA receptor from M. tuberculosis. Based on the receptor residues-ligand distances we preprocessed all docking results to generate appropriate input to mine data. Data preprocessing was done by calculating the shortest interatomic distances between the ligand and the receptor’s residues for each docking result. They were the predictive attributes. The target attribute was the estimated free-energy of binding (FEB) value calculated by the AutodDock3.0.5 software. The mining inputs were submitted to the M5P model tree algorithm. It resulted in short and understandable trees. On the basis of the correlation values, for NADH, TCL and PIF we obtained more than 95% correlation while for ETH, only about 60%. Post processing the generated model trees for each of its linear models (LMs), we calculated the average FEB for their associated instances. From these values we considered a LM as representative if its average FEB was smaller than or equal the average FEB of the test set. The instances in the selected LMs were considered the most promising snapshots. It totalized 1,521, 1,780, 2,085 and 902 snapshots, for NADH, TCL, PIF and ETH respectively.ConclusionsBy post processing the generated model trees we were able to propose a criterion of selection of linear models which, in turn, is capable of selecting a set of promising receptor conformations. As future work we intend to go further and use these results to elaborate a strategy to preprocess the receptors 3-D spatial conformation in order to predict FEB values. Besides, we intend to select other compounds, among the million catalogued, that may be promising as new drug candidates for our particular protein receptor target.

FReDoWS: a method to automate molecular docking simulations with explicit receptor flexibility and snapshots selection.

BackgroundIn silico molecular docking is an essential step in modern drug discovery when driven by a well defined macromolecular target. Hence, the process is called structure-based or rational drug design (RDD). In the docking step of RDD the macromolecule or receptor is usually considered a rigid body. However, we know from biology that macromolecules such as enzymes and membrane receptors are inherently flexible. Accounting for this flexibility in molecular docking experiments is not trivial. One possibility, which we call a fully-flexible receptor model, is to use a molecular dynamics simulation trajectory of the receptor to simulate its explicit flexibility. To benefit from this concept, which has been known since 2000, it is essential to develop and improve new tools that enable molecular docking simulations of fully-flexible receptor models.ResultsWe have developed a Flexible-Receptor Docking Workflow System (FReDoWS) to automate molecular docking simulations using a fully-flexible receptor model. In addition, it includes a snapshot selection feature to facilitate acceleration the virtual screening of ligands for well defined disease targets. FReDoWS usefulness is demonstrated by investigating the docking of four different ligands to flexible models of Mycobacterium tuberculosis’ wild type InhA enzyme and mutants I21V and I16T. We find that all four ligands bind effectively to this receptor as expected from the literature on similar, but wet experiments.ConclusionsA work that would usually need the manual execution of many computer programs, and the manipulation of thousands of files, was efficiently and automatically performed by FReDoWS. Its friendly interface allows the user to change the docking and execution parameters. Besides, the snapshot selection feature allowed the acceleration of docking simulations. We expect FReDoWS to help us explore more of the role flexibility plays in receptor-ligand interactions. FReDoWS can be made available upon request to the authors.

Automatic design of decision-tree induction algorithms tailored to flexible-receptor docking data

BackgroundThis paper addresses the prediction of the free energy of binding of a drug candidate with enzyme InhA associated with Mycobacterium tuberculosis. This problem is found within rational drug design, where interactions between drug candidates and target proteins are verified through molecular docking simulations. In this application, it is important not only to correctly predict the free energy of binding, but also to provide a comprehensible model that could be validated by a domain specialist. Decision-tree induction algorithms have been successfully used in drug-design related applications, specially considering that decision trees are simple to understand, interpret, and validate. There are several decision-tree induction algorithms available for general-use, but each one has a bias that makes it more suitable for a particular data distribution. In this article, we propose and investigate the automatic design of decision-tree induction algorithms tailored to particular drug-enzyme binding data sets. We investigate the performance of our new method for evaluating binding conformations of different drug candidates to InhA, and we analyze our findings with respect to decision tree accuracy, comprehensibility, and biological relevance.ResultsThe empirical analysis indicates that our method is capable of automatically generating decision-tree induction algorithms that significantly outperform the traditional C4.5 algorithm with respect to both accuracy and comprehensibility. In addition, we provide the biological interpretation of the rules generated by our approach, reinforcing the importance of comprehensible predictive models in this particular bioinformatics application.ConclusionsWe conclude that automatically designing a decision-tree algorithm tailored to molecular docking data is a promising alternative for the prediction of the free energy from the binding of a drug candidate with a flexible-receptor.

Effect of the explicit flexibility of the InhA enzyme from Mycobacterium tuberculosis in molecular docking simulations

BackgroundProtein/receptor explicit flexibility has recently become an important feature of molecular docking simulations. Taking the flexibility into account brings the docking simulation closer to the receptors’ real behaviour in its natural environment. Several approaches have been developed to address this problem. Among them, modelling the full flexibility as an ensemble of snapshots derived from a molecular dynamics simulation (MD) of the receptor has proved very promising. Despite its potential, however, only a few studies have employed this method to probe its effect in molecular docking simulations. We hereby use ensembles of snapshots obtained from three different MD simulations of the InhA enzyme from M. tuberculosis (Mtb), the wild-type (InhA_wt), InhA_I16T, and InhA_I21V mutants to model their explicit flexibility, and to systematically explore their effect in docking simulations with three different InhA inhibitors, namely, ethionamide (ETH), triclosan (TCL), and pentacyano(isoniazid)ferrate(II) (PIF).ResultsThe use of fully-flexible receptor (FFR) models of InhA_wt, InhA_I16T, and InhA_I21V mutants in docking simulation with the inhibitors ETH, TCL, and PIF revealed significant differences in the way they interact as compared to the rigid, InhA crystal structure (PDB ID: 1ENY). In the latter, only up to five receptor residues interact with the three different ligands. Conversely, in the FFR models this number grows up to an astonishing 80 different residues. The comparison between the rigid crystal structure and the FFR models showed that the inclusion of explicit flexibility, despite the limitations of the FFR models employed in this study, accounts in a substantial manner to the induced fit expected when a protein/receptor and ligand approach each other to interact in the most favourable manner.ConclusionsProtein/receptor explicit flexibility, or FFR models, represented as an ensemble of MD simulation snapshots, can lead to a more realistic representation of the induced fit effect expected in the encounter and proper docking of receptors to ligands. The FFR models of InhA explicitly characterizes the overall movements of the amino acid residues in helices, strands, loops, and turns, allowing the ligand to properly accommodate itself in the receptor’s binding site. Utilization of the intrinsic flexibility of Mtb’s InhA enzyme and its mutants in virtual screening via molecular docking simulation may provide a novel platform to guide the rational or dynamical-structure-based drug design of novel inhibitors for Mtb’s InhA. We have produced a short video sequence of each ligand (ETH, TCL and PIF) docked to the FFR models of InhA_wt. These videos are available at http://www.inf.pucrs.br/~osmarns/LABIO/Videos_Cohen_et_al_19_07_2011.htm.

Automating molecular docking with explicit receptor flexibility using scientific workflows

Computer assisted drug design (CADD) is a process involving the execution of many computer programs, ensuring that the ligand binds optimally to its receptor. This process is usually executed using shell scripts which input parameters assignments and result analyses are complex and time consuming. Moreover, receptors and ligands are naturally flexible molecules. In order to explicitly model the receptor flexibility during molecular docking experiments, we propose to use different receptor conformations derived from a molecular dynamics simulation trajectory. This work presents an integrated scientific workflow solution aiming at automating molecular docking with explicit inclusion of receptor flexibility. Enhydra JAWE and Shark software tools were used to model and execute workflows, respectively. To test our approach we performed docking experiments with the M. tuberculosis enzyme InhA (receptor) and three ligands: NADH, IPCF and TCL. The results illustrate the effectiveness of both the proposed workflow and the implementation of the docking processes.

FReDD: Supporting Mining Strategies through a Flexible-Receptor Docking Database

Among different alternatives to consider the receptor flexibility in molecular docking experiments we opt to execute a series of docking using receptor snapshots generated by molecular dynamics simulations. Our target is the InhA enzyme from Mycobacterium tuberculosis bound to NADH, TCL, PIF and ETH ligands. After testing some mining strategies on these data, we conclude that, to obtain better outcomes, the development of an organized repository is especially useful. Thus, we built a comprehensive and robust database called FReDD to store the InhA-ligand docking results. Using this database we concentrate efforts on data mining to explore the docking results in order to accelerate the identification of promising ligands against the InhA target.

Mining flexible‐receptor molecular docking data

Knowledge discovery in databases has become an integral part of practically every aspect of bioinformatics research, which usually produces, and has to process, very large amounts of data. Rational drug design is one of the current scientific areas that has greatly benefited from bioinformatics, particularly a step, which analyzes receptor–ligand interactions via molecular docking simulations. An important challenge is the inclusion of the receptor flexibility since they can become computationally very demanding. We have represented this explicit flexibility as a series of different conformations derived from a molecular dynamics simulation trajectory of the receptor. This model has been termed as the fully flexible receptor (FFR) model. In our studies, the receptor is the enzyme InhA from Mycobacterium tuberculosis, which is the major drug target for the treatment of tuberculosis. The FFR model of InhA (named FFR_InhA) was docked to four ligands, namely, nicotinamide adenine dinucleotide, pentacyano(isoniazid)ferrate II, triclosan, and ethionamide, thus, generating very large amounts of data, which needs to be mined to produce useful knowledge to help accelerate drug discovery and development. Very little work has been done in this area. In this article, we review our work on the application of classification decision trees, regression model tree, and association rules using properly preprocessed data of the FFR molecular docking results, and show how they can provide an improved understanding of the FFR_InhA‐ligand behavior. Furthermore, we explain how data mining techniques can support the acceleration of molecular docking simulations of FFR models.

Predicting the binding properties of single walled carbon nanotubes (SWCNT) with an ADP/ATP mitochondrial carrier using molecular docking, chemoinformatics, and nano-QSBR perturbation theory

Interactions between the single walled carbon nanotube (SWCNT) family and a mitochondrial ADP/ATP carrier (ANT-1) were evaluated using constitutional (functional groups, number of carbon atoms, etc.) and electronic nanodescriptors defined by (n, m)-Hamada indexes (armchair, zig-zag and chiral). The Free Energy of Binding (FEB) was determined by molecular docking simulation and the results showed that FEB was statistically more negative (p SWCNT-OH > SWCNT, suggesting that polar groups favor the anchorage to ANT-1. In this regard, it was showed that key ANT-1 amino acids (Arg 79, Asn 87, Lys 91, Arg 187, Arg 234 and Arg 279) responsible for ADP-transport were conserved in ANT-1 from different species examined to predict SWCNT interactions, including shrimp Litopenaeus vannamei and fish Danio rerio commonly employed in ecotoxicology. The SWCNT-ANT-1 inter-atomic distances for the key ANT-1 amino acids were similar to that with carboxyatractyloside, a classical inhibitor of ANT-1. Significant linear relationships between FEB and n-Hamada index were found for zig-zag SWCNT and SWCNT-COOH (R2 = 0.95 in both cases). A Perturbation Theory-Nano-Quantitative Structure-Binding Relationship (PT-NQSBR) model was fitted that was able to distinguish between strong (FEB < −14.7 kcal mol−1) and weak (FEB ≥ −14.7 kcal mol−1) SWCNT–ANT-1 interactions. A simple ANT-1-inhibition respiratory assay employing mitochondria suspension from L. vannamei, showed good accordance with the predicted model. These results indicate that this methodology can be employed in massive virtual screenings and used for making regulatory decisions in nanotoxicology.

A methodology for selecting the most suitable cluster validation internal indices

Validation of clustering results is an important issue in the context of machine learning research and it is essential for the success of clustering applications. Choosing the appropriate validation index for evaluating the results of a particular clustering algorithm remains a challenge. The quality of partitions generated by different clustering algorithms can be evaluated using different indices based on external or internal criteria. In this paper, we have proposed a methodology for selecting the most suitable cluster validation internal index, relating external and internal criteria through a regression model applied on the results of partitioning clustering algorithm.

Context-based preprocessing of molecular docking data

BackgroundData preprocessing is a major step in data mining. In data preprocessing, several known techniques can be applied, or new ones developed, to improve data quality such that the mining results become more accurate and intelligible. Bioinformatics is one area with a high demand for generation of comprehensive models from large datasets. In this article, we propose a context-based data preprocessing approach to mine data from molecular docking simulation results. The test cases used a fully-flexible receptor (FFR) model of Mycobacterium tuberculosis InhA enzyme (FFR_InhA) and four different ligands.ResultsWe generated an initial set of attributes as well as their respective instances. To improve this initial set, we applied two selection strategies. The first was based on our context-based approach while the second used the CFS (Correlation-based Feature Selection) machine learning algorithm. Additionally, we produced an extra dataset containing features selected by combining our context strategy and the CFS algorithm. To demonstrate the effectiveness of the proposed method, we evaluated its performance based on various predictive (RMSE, MAE, Correlation, and Nodes) and context (Precision, Recall and FScore) measures.ConclusionsStatistical analysis of the results shows that the proposed context-based data preprocessing approach significantly improves predictive and context measures and outperforms the CFS algorithm. Context-based data preprocessing improves mining results by producing superior interpretable models, which makes it well-suited for practical applications in molecular docking simulations using FFR models.