http://www.jcheminf.com The latest research articles published by Journal of Cheminformatics 2010-03-11T00:00:00Z This is an RSS newsfeed from BioMed Central It is intended to be used with an RSS reader. For more information about RSS newsfeeds from BioMed Central, visit http://www.biomedcentral.com/info/about/rss/ Background: The virtual screening of large compound databases is an important application of structural-activity relationship models. Due to the high structural diversity of these data sets, it is impossible for machine learning based QSAR models, which rely on a specific training set, to give reliable results for all compounds. Thus, it is important to consider the subset of the chemical space in which the model is applicable. The approaches to this problem that have been published so far mostly use vectorial descriptor representations to define this domain of applicability of the model. Unfortunately, these cannot be extended easily to structured kernel-based machine learning models. For this reason, we propose three approaches to estimate the domain of applicability of a kernel-based QSAR model. Results: We evaluated three kernel-based applicability domain estimations using three different structured kernels on three virtual screening tasks. Each experiment consisted of the training of a kernel-based QSAR model using support vector regression and the ranking of a disjoint screening data set according to the predicted activity. For each prediction, the applicability of the model for the respective compound is quantitatively described using a score obtained by an applicability domain formulation. The suitability of the applicability domain estimation is evaluated by comparing the model performance on the subsets of the screening data sets obtained by different thresholds for the applicability scores. This comparison indicates that it is possible to separate the part of the chemspace, in which the model gives reliable predictions, from the part consisting of structures too dissimilar to the training set to apply the model successfully. A closer inspection reveals that the virtual screening performance of the model is considerably improved if half of the molecules, those with the lowest applicability scores, are omitted from the screening. Conclusion: The proposed applicability domain formulations for kernel-based QSAR models can successfully identify compounds for which no reliable predictions can be expected from the model. The resulting reduction of the search space and the elimination of some of the active compounds should not be considered as a drawback, because the results indicate that, in most cases, these omitted ligands would not be found by the model anyway. http://www.jcheminf.com/content/2/1/2 Nikolas Fechner Andreas Jahn Georg Hinselmann Andreas Zell Journal of Cheminformatics 2010, 2:2 2010-03-11T00:00:00Z doi:10.1186/1758-2946-2-2 Journal of Cheminformatics 1758-2946 2 2 2010-03-11T00:00:00Z PDF A molecule editor, that is program for input and editing of molecules, is an indispensable part of every cheminformatics or molecular processing system. This review focuses on a special type of molecule editors, namely those that are used for molecule structure input on the web. Scientific computing is now moving more and more in the direction of web services and cloud computing, with servers scattered all around the Internet. Thus a web browser has become the universal scientific user interface, and a tool to edit molecules directly within the web browser is essential.The review covers a history of web-based structure input, starting with simple text entry boxes and early molecule editors based on clickable maps, before moving to the current situation dominated by Java applets. One typical example - the popular JME Molecule Editor - will be described in more detail. Modern Ajax server-side molecule editors are also presented. And finally, the possible future direction of web-based molecule editing, based on technologies like JavaScript and Flash, is discussed. http://www.jcheminf.com/content/2/1/1 Peter Ertl Journal of Cheminformatics 2010, 2:1 2010-02-02T00:00:00Z doi:10.1186/1758-2946-2-1 Journal of Cheminformatics 1758-2946 2 1 2010-02-02T00:00:00Z XML Background: The topological maximum cross correlation (TMACC) descriptors are alignment-independent 2D descriptors for the derivation of QSARs. TMACC descriptors are generated using atomic properties determined by molecular topology. Previous validation (J Chem Inf Model 2007, 47: 626-634) of the TMACC descriptor suggests it is competitive with the current state of the art. Results: Here, we illustrate the interpretability of the TMACC descriptors, through the analysis of the QSARs of inhibitors of angiotensin converting enzyme (ACE) and dihydrofolate reductase (DHFR). In the case of the ACE inhibitors, the TMACC interpretation shows features specific to C-domain inhibition, which have not been explicitly identified in previous QSAR studies. Conclusions: The TMACC interpretation can provide new insight into the structure-activity relationships studied. Freely available, open source software for generating the TMACC descriptors can be downloaded from http://comp.chem.nottingham.ac.uk. http://www.jcheminf.com/content/1/1/22 Benson Spowage Craig Bruce Jonathan Hirst Journal of Cheminformatics 2009, 1:22 2009-12-24T00:00:00Z doi:10.1186/1758-2946-1-22 Journal of Cheminformatics 1758-2946 1 22 2009-12-24T00:00:00Z XML Background: There are three main problems associated with the virtual screening of bioassay data. The first is access to freely-available curated data, the second is the number of false positives that occur in the physical primary screening process, and finally the data is highly-imbalanced with a low ratio of Active compounds to Inactive compounds. This paper first discusses these three problems and then a selection of Weka cost-sensitive classifiers (Naive Bayes, SVM, C4.5 and Random Forest) are applied to a variety of bioassay datasets. Results: Pharmaceutical bioassay data is not readily available to the academic community. The data held at PubChem is not curated and there is a lack of detailed cross-referencing between Primary and Confirmatory screening assays. With regard to the number of false positives that occur in the primary screening process, the analysis carried out has been shallow due to the lack of cross-referencing mentioned above. In six cases found, the average percentage of false positives from the High-Throughput Primary screen is quite high at 64%. For the cost-sensitive classification, Weka's implementations of the Support Vector Machine and C4.5 decision tree learner have performed relatively well. It was also found, that the setting of the Weka cost matrix is dependent on the base classifier used and not solely on the ratio of class imbalance. Conclusions: Understandably, pharmaceutical data is hard to obtain. However, it would be beneficial to both the pharmaceutical industry and to academics for curated primary screening and corresponding confirmatory data to be provided. Two benefits could be gained by employing virtual screening techniques to bioassay data. First, by reducing the search space of compounds to be screened and secondly, by analysing the false positives that occur in the primary screening process, the technology may be improved. The number of false positives arising from primary screening leads to the issue of whether this type of data should be used for virtual screening. Care when using Weka's cost-sensitive classifiers is needed - across the board misclassification costs based on class ratios should not be used when comparing differing classifiers for the same dataset. http://www.jcheminf.com/content/1/1/21 Amanda Schierz Journal of Cheminformatics 2009, 1:21 2009-12-22T00:00:00Z doi:10.1186/1758-2946-1-21 Journal of Cheminformatics 1758-2946 1 21 2009-12-22T00:00:00Z XML PubChem is an important public, Web-based information source for chemical and bioactivity information. In order to provide convenient structure search methods on compounds stored in this database, one mandatory component is a Web-based drawing tool for interactive sketching of chemical query structures. Web-enabled chemical structure sketchers are not new, being in existence for years; however, solutions available rely on complex technology like Java applets or platform-dependent plug-ins. Due to general policy and support incident rate considerations, Java-based or platform-specific sketchers cannot be deployed as a part of public NCBI Web services. Our solution: a chemical structure sketching tool based exclusively on CGI server processing, client-side JavaScript functions, and image sequence streaming. The PubChem structure editor does not require the presence of any specific runtime support libraries or browser configurations on the client. It is completely platform-independent and verified to work on all major Web browsers, including older ones without support for Web2.0 JavaScript objects. http://www.jcheminf.com/content/1/1/20 Wolf-D. Ihlenfeldt Evan Bolton Stephen Bryant Journal of Cheminformatics 2009, 1:20 2009-12-17T00:00:00Z doi:10.1186/1758-2946-1-20 Journal of Cheminformatics 1758-2946 1 20 2009-12-17T00:00:00Z XML Background: The identification of promising drug leads from a large database of compounds is an important step in the preliminary stages of drug design. Although shape is known to play a key role in the molecular recognition process, its application to virtual screening poses significant hurdles both in terms of the encoding scheme and speed. Results: In this study, we have examined the efficacy of the alignment independent three-dimensional Zernike descriptor (3DZD) for fast shape based similarity searching. Performance of this approach was compared with several other methods including the statistical moments based ultrafast shape recognition scheme (USR) and SIMCOMP, a graph matching algorithm that compares atom environments. Three benchmark datasets are used to thoroughly test the methods in terms of their ability for molecular classification, retrieval rate, and performance under the situation that simulates actual virtual screening tasks over a large pharmaceutical database. The 3DZD performed better than or comparable to the other methods examined, depending on the datasets and evaluation metrics used. Reasons for the success and the failure of the shape based methods for specific cases are investigated. Based on the results for the three datasets, general conclusions are drawn with regard to their efficiency and applicability. Conclusion: The 3DZD has unique ability for fast comparison of three-dimensional shape of compounds. Examples analyzed illustrate the advantages and the room for improvements for the 3DZD. http://www.jcheminf.com/content/1/1/19 Vishwesh Venkatraman Padmasini Ramji Chakravarthy Daisuke Kihara Journal of Cheminformatics 2009, 1:19 2009-12-17T00:00:00Z doi:10.1186/1758-2946-1-19 Journal of Cheminformatics 1758-2946 1 19 2009-12-17T00:00:00Z XML Background: Modern chemistry laboratories operate with a wide range of software applications under different operating systems, such as Windows, LINUX or Mac OS X. Instead of installing software on different computers it is possible to install those applications on a single computer using Virtual Machine software. Software platform virtualization allows a single guest operating system to execute multiple other operating systems on the same computer. We apply and discuss the use of virtual machines in chemistry research and teaching laboratories. Results: Virtual machines are commonly used for cheminformatics software development and testing. Benchmarking multiple chemistry software packages we have confirmed that the computational speed penalty for using virtual machines is low and around 5% to 10%. Software virtualization in a teaching environment allows faster deployment and easy use of commercial and open source software in hands-on computer teaching labs. Conclusion: Software virtualization in chemistry, mass spectrometry and cheminformatics is needed for software testing and development of software for different operating systems. In order to obtain maximum performance the virtualization software should be multi-core enabled and allow the use of multiprocessor configurations in the virtual machine environment. Server consolidation, by running multiple tasks and operating systems on a single physical machine, can lead to lower maintenance and hardware costs especially in small research labs. The use of virtual machines can prevent software virus infections and security breaches when used as a sandbox system for internet access and software testing. Complex software setups can be created with virtual machines and are easily deployed later to multiple computers for hands-on teaching classes. We discuss the popularity of bioinformatics compared to cheminformatics as well as the missing cheminformatics education at universities worldwide. http://www.jcheminf.com/content/1/1/18 Tobias Kind Tim Leamy Julie Leary Oliver Fiehn Journal of Cheminformatics 2009, 1:18 2009-11-16T00:00:00Z doi:10.1186/1758-2946-1-18 Journal of Cheminformatics 1758-2946 1 18 2009-11-16T00:00:00Z XML Background: Registration, indexing and searching of chemical structures in relational databases is one of the core areas of cheminformatics. However, little detail has been published on the inner workings of search engines and their development has been mostly closed-source. We decided to develop an open source chemistry extension for Oracle, the de facto database platform in the commercial world. Results: Here we present OrChem, an extension for the Oracle 11G database that adds registration and indexing of chemical structures to support fast substructure and similarity searching. The cheminformatics functionality is provided by the Chemistry Development Kit. OrChem provides similarity searching with response times in the order of seconds for databases with millions of compounds, depending on a given similarity cut-off. For substructure searching, it can make use of multiple processor cores on today's powerful database servers to provide fast response times in equally large data sets.AvailabilityOrChem is free software and can be redistributed and/or modified under the terms of the GNU Lesser General Public License as published by the Free Software Foundation. All software is available via http://orchem.sourceforge.net. http://www.jcheminf.com/content/1/1/17 Mark Rijnbeek Christoph Steinbeck Journal of Cheminformatics 2009, 1:17 2009-10-22T00:00:00Z doi:10.1186/1758-2946-1-17 Journal of Cheminformatics 1758-2946 1 17 2009-10-22T00:00:00Z XML Background: Detailed experimental three dimensional structures of carbohydrates are often difficult to acquire. Molecular modelling and computational conformation prediction are therefore commonly used tools for three dimensional structure studies. Modelling procedures generally require significant training and computing resources, which is often impractical for most experimental chemists and biologists. Shape has been developed to improve the availability of modelling in this field. Results: The Shape software package has been developed for simplicity of use and conformation prediction performance. A trivial user interface coupled to an efficient genetic algorithm conformation search makes it a powerful tool for automated modelling. Carbohydrates up to a few hundred atoms in size can be investigated on common computer hardware. It has been shown to perform well for the prediction of over four hundred bioactive oligosaccharides, as well as compare favourably with previously published studies on carbohydrate conformation prediction. Conclusion: The Shape fully automated conformation prediction can be used by scientists who lack significant modelling training, and performs well on computing hardware such as laptops and desktops. It can also be deployed on computer clusters for increased capacity. The prediction accuracy under the default settings is good, as it agrees well with experimental data and previously published conformation prediction studies. This software is available both as open source and under commercial licenses. http://www.jcheminf.com/content/1/1/16 Jimmy Rosen Laurence Miguet Serge Perez Journal of Cheminformatics 2009, 1:16 2009-09-21T00:00:00Z doi:10.1186/1758-2946-1-16 Journal of Cheminformatics 1758-2946 1 16 2009-09-21T00:00:00Z XML Background: Molecular docking methods are commonly used for predicting binding modes and energies of ligands to proteins. For accurate complex geometry and binding energy estimation, an appropriate method for calculating partial charges is essential. AutoDockTools software, the interface for preparing input files for one of the most widely used docking programs AutoDock 4, utilizes the Gasteiger partial charge calculation method for both protein and ligand charge calculation. However, it has already been shown that more accurate partial charge calculation - and as a consequence, more accurate docking- can be achieved by using quantum chemical methods. For docking calculations quantum chemical partial charge calculation as a routine was only used for ligands so far. The newly developed Mozyme function of MOPAC2009 allows fast partial charge calculation of proteins by quantum mechanical semi-empirical methods. Thus, in the current study, the effect of semi-empirical quantum-mechanical partial charge calculation on docking accuracy could be investigated. Results: The docking accuracy of AutoDock 4 using the original AutoDock scoring function was investigated on a set of 53 protein ligand complexes using Gasteiger and PM6 partial charge calculation methods. This has enabled us to compare the effect of the partial charge calculation method on docking accuracy utilizing AutoDock 4 software. Our results showed that the docking accuracy in regard to complex geometry (docking result defined as accurate when the RMSD of the first rank docking result complex is within 2 Å of the experimentally determined X-ray structure) significantly increased when partial charges of the ligands and proteins were calculated with the semi-empirical PM6 method.Out of the 53 complexes analyzed in the course of our study, the geometry of 42 complexes were accurately calculated using PM6 partial charges, while the use of Gasteiger charges resulted in only 28 accurate geometries. The binding affinity estimation was not influenced by the partial charge calculation method - for more accurate binding affinity prediction development of a new scoring function for AutoDock is needed. Conclusion: Our results demonstrate that the accuracy of determination of complex geometry using AutoDock 4 for docking calculation greatly increases with the use of quantum chemical partial charge calculation on both the ligands and proteins. http://www.jcheminf.com/content/1/1/15 Zsolt Bikadi Eszter Hazai Journal of Cheminformatics 2009, 1:15 2009-09-11T00:00:00Z doi:10.1186/1758-2946-1-15 Journal of Cheminformatics 1758-2946 1 15 2009-09-11T00:00:00Z XML