Abstracts


[Back to top]
How to Efficiently Include Receptor Flexibility During Computational Docking
Andreas May, Florian Sieker and Martin Zacharias

Target-based drug design uses available 3D structural information of receptor molecules to either dock putative ligand molecules to receptor binding sites or to de-novo design new ligands. In many cases accurate prediction of putative binding geometries requires the appropriate inclusion of conformational flexibility of both the ligand as well as the receptor structure. The problem of appropriate treatment of conformational flexibility during docking is also tightly connected to the improvement of scoring a docked ligand-receptor complex. Highly accurate scoring of a ligand placement is only possible if the complex geometry has been predicted with high precision. Considerable progress has already been achieved in modeling the conformational flexibility of small organic ligands during docking. Although of similar importance, receptor flexibility has not been tackled satisfactorily despite the steady increase in computational power. Especially during virtual screening of large drug-like compound libraries the target structure often is still kept rigid. However, many protein structures undergo local structural changes (side chain or loop motions) as well as global changes in the backbone geometry upon complex formation. In recent years, several promising approaches to efficiently tackle receptor flexibility have been introduced ranging from conformational ensemble methods to explicit inclusion of the most relevant receptor degrees of freedom. Possible applications and future directions on improving flexible docking approaches will be discussed.


[Back to top]
Predictive QSAR Models for Polyspecific Drug Targets: The Importance of Feature Selection
M.A. Demel, Andres G.K. Janecek, Khac-Minh Thai, Gerhard F. Ecker and Wilfried N. Gansterer

Since the advent of QSAR (quantitative structure activity relationship) modeling quantitative representations of molecular structures are encoded in terms of information-preserving descriptor values. Nowadays, a nearly infinite variety of potential descriptors is available and descriptor selection is no longer a task which can be done manually. There is an increasing need for automation in order to reduce the dimensionality of the descriptor space. Classical feature selection (FS) and dimensionality reduction (DR) methods like principal component analysis, which relies on the selection of those descriptors that contribute most to the variance of a data set, often fail in providing the best classification result. More sophisticated methods like genetic algorithms, self-organizing-maps and stepwise linear discriminant analysis have proven to be useful techniques in the process of selecting descriptors with a significant discriminative power. The topic FS and DR becomes even more important when predictive models are approached which should describe the QSAR of highly promiscuous target proteins. The ABC-transporter family, the cardiac hERG-potassium channel, and the hepatic cytochrom-P450-family are classical representatives of such poly-specific proteins. In this case the interaction pattern is a rather complex one and thus the selection of the most predictive descriptors needs advanced methods. This review surveys FS and DR methods that have recently been successfully applied to classify ligands of poly-specific target proteins.


[Back to top]
How to Measure the Similarity Between Protein Ligand-Binding Sites?
Esther Kellenberger, Claire Schalon and Didier Rognan

Quantification of local similarity between protein 3D structures is a promising tool in computer-aided drug design and prediction of biological function. Over the last ten years, several computational methods were proposed, mostly based on geometrical comparisons. This review summarizes the recent literature and gives an overview of available programs. A particular interest is given to the underlying methodologies. Our analysis points out strengths and weaknesses of the various approaches. If all described methods work relatively well when two binding sites obviously resemble each other, scoring potential solutions remains a difficult issue, especially if the similarity is low. The other challenging question is the protein flexibility, which is indeed difficult to evaluate from a static representation. Last, most of recently developed techniques are fast and can be applied to large amounts of data. Examples were carefully chosen to illustrate the wide applicability domain of the most popular methods: detection of common structural motifs, identification of secondary targets for a drug-like compound, comparison of binding sites across a functional family, comparison of homology models, database screening.


[Back to top]
Docking and High Throughput Docking: Successes and the Challenge of Protein Flexibility
Claudio N. Cavasotto and Narender Singh

Protein structure-based approaches in lead optimization and in silico screening of chemical libraries based on ligand docking have increasingly become part of many drug discovery projects, mainly due to the technical improvements in crystallography, the support of modern software, and the ever increasing computational power. The use of three-dimensional structural information of therapeutic targets has long been recognized to initiate and accelerate many drug design programs in the past, since it offers the possibility of finding novel scaffolds, different from the existing active compounds. In spite of its many successes, structure-based virtual screening or high throughput docking still has several limitations at the methodological level, not the least of which is protein flexibility, ignored by most of the docking programs, which treat the receptor as a rigid entity. This may impact the accuracy of virtual screening at the docking and at the scoring level. The authors will first present the latest successful stories in high throughput docking. After reviewing the concepts of protein dynamics and binding, and its impact in ligand docking, the current approaches to incorporate protein mobility in docking-based virtual screening will be presented and discussed.


[Back to top]
Quantitative Sequence-Activity Model (QSAM): Applying QSAR Strategy to Model and Predict Bioactivity and Function of Peptides, Proteins and Nucleic Acids
Peng Zhou, Feifei Tian, Yuqian Wu, Zhiliang Li and Zhicai Shang

Traditional quantitative structure-activity relationship (QSAR) is a term describing a variety of approaches that are of substantial interest for chemistry. Quantitative sequence-activity model (QSAM), applying QSAR strategy to explore sequence-activity/function relationship for biosystems, is greatly meaningful but meanwhile extremely difficult. For biomolecules, high molecular weight, diverse structural morphology and intricate interaction network all bring in traditional QSAR methodologies unprecedented challenges. This article comprehensively reviewed developing process, current state and future perspective of QSAM, concerning its applications into fields of pharmacy, food science, immunology and molecular biology. Besides, discipline-crossing and amalgamation of QSAM with QSAR, bioinformatics and computational biology were also discussed.


[Back to top]
Visualization of the Chemical Space in Drug Discovery
Jose L. Medina-Franco, Karina Martínez-Mayorga, Marc A. Giulianotti, Richard A. Houghten and Clemencia Pinilla

Chemical space has become a key concept in drug discovery. The continued growth in the number of molecules available raises the question regarding how many compounds may exist and which ones have the potential to become drugs. Analysis and visualization of the chemical space covered by public, commercial, in-house and virtual compound collections have found multiple applications in diversity analysis, in silicoproperty profiling, data mining, virtual screening, library design, prioritization in screening campaigns, and acquisition of compound collections, among others. This review covers several techniques, computational programs and approaches that have been developed to visualize, navigate and study the chemical space of molecular databases. Techniques developed in our group are presented including a quantitative assessment of the multi-fusion similarity maps. Additionally an application of 3D-similarity, based on the overlay of chemical structures, to represent the chemical space is introduced. Several comparisons of the chemical space covered by compound collections from different sources such as combinatorial libraries, drugs and natural products, or directed to specific therapeutic areas are also discussed.


[Back to top]
Novel Quantitative Structure-Activity Studies of HIV-1 Protease Inhibitors of the Cyclic Urea Type Using Descriptors Derived from Molecular Dynamics and Molecular Orbital Calculations
Tatsusada Yoshida, Toshio Fujita and Hiroshi Chuman

Human immunodeficiency virus type 1 protease (HIV-1 PR) is an essential enzyme for the replication cycle of HIV-1. HIV-1 PR inhibitors have been extensively investigated as anti-AIDS drugs. For developments of HIV-1 PR inhibitors more promising than those utilized at the moment, the construction of reliable QSAR models that can elucidate the inhibitory mechanism as consistently as possible should be one of the most significant issues. Garg, Kurup, and their groups published comprehensive QSAR studies using past structure-activity data for HIV-1 PR inhibitors, and summarized some physicochemical structural factors of inhibitors that govern variations in the inhibitory activity for various structural types. There seem to exist much to be clarified further, especially for effects of electronic structure of inhibitors. It is also expected to incorporate structural and physicochemical information about the enzyme protein into the QSAR model. In this article, we reviewed our own QSAR study on a series of cyclic urea inhibitors with newly proposed QSAR descriptors. We performed molecular dynamics simulations of HIV-1 PR-inhibitor complexes to provide the accurate geometry to the fragment molecular orbital (FMO) calculations as well as to the estimation of an accessible surface area descriptor for inhibitors and amino acid residues. With the FMO procedure to cover full electronic feature of three-dimensional structure of protease-inhibitor complexes, we derived electronic descriptors for inhibitors and amino acid residues. The successful results are believed to provide a new insight into QSAR and understanding of binding mechanism of inhibitors with HIV-1 PR at atomic and electronic levels.


[Back to top]
Multimode Methods Applied on MIA Descriptors in QSAR
Matheus P. Freitas, Elaine F.F. da Cunha, Teodorico C. Ramalho and Mohammad Goodarzi

Since the introduction of physicochemical descriptors to derive useful QSAR (quantitative structure-activity relationship) models, some regression methods have been applied to linearly correlate dependent (bioactivities) and independent variables. Multiple linear regression (MLR) has been widely used when the number of samples (rows) exceed the amount of descriptors (columns), whilst partial least squares (PLS) is the most commonly applied regression method in 3D QSAR (e.g. CoMFA and related methods), where a large number of descriptors are generated. The recently implemented MIA-QSAR (Multivariate Image Analysis applied to QSAR) method is a especial (not only) case in which the descriptors (pixels) for each active compound result in a three-way array after grouping samples to give a data set. Such array may be properly treated by using N-way methods, such as multilinear PLS (N-PLS) and parallel factor analysis (PARAFAC). However, these methods have not been appropriately explored in QSAR studies, despite their supposed advantages over well established methods. Thus, this review formally details the MIA-QSAR approach prior to presenting two promising multimode methods to be applied on MIA descriptors, namely N-PLS and PARAFAC. Also, the suitability of such methods is discussed in terms of application to a case study (a series of anti-HIV compounds) and comparison to traditional (bilinear) PLS and docking studies.


[Back to top]
Ligand-Based Approaches in Virtual Screening
Dominique Douguet

Although there are many more receptor structures than there were in the 1970s and 1980s, drug discovery remains dominated by empirical screening and substrate-based drug design. Computer-aided drug design methods have become value-adding disciplines that now contribute to the early stage of the drug discovery process [1, 2]. Computational methods encompass all aspects of drug discovery from target assessment to lead optimization. The computational strategy varies from case to case and can be influenced by several situational variables: lead hunting or lead optimization, requirement for a novel lead class, type of biological assay, structural information available, known classes of ligands, allocated chemistry resources… Today, drug discovery is still a complex and approximate science. Thus, incorporating knowledge-based approaches like ligand-based screenings may bias the process towards success. This review describes these strategies with practical applications and presents future perspectives of ligand-based screening.


[Back to top]
ISIDA - Platform for Virtual Screening Based on Fragment and Pharmacophoric Descriptors
Alexandre Varnek, Denis Fourches, Dragos Horvath, Olga Klimchuk, Cedric Gaudin, Philippe Vayer, Vitaly Solov’ev, Frank Hoonakker, Igor V. Tetko and Gilles Marcou

In this paper we illustrate the application of the ISIDA (In SIlico design and Data Analysis) software to perform virtual screening of large databases of compounds and reactions and to assess some ADME/Tox properties. ISIDA represents an ensemble of tools allowing users to store, search and analyze the data, to perform similarity searches in large databases of molecules and reactions, to build and validate QSAR models, and to generate and screen virtual combinatorial libraries. It uses its own descriptors (substructural molecular fragments and fuzzy pharmacophore triplets). Workflow can be easily organized by combining different ISIDA modules. Several examples of ISIDA applications (similarity search of potent benzodiazepine ligands with FPT, QSAR modeling of aqueous solubility, aquatic toxicity, tissue-air partition coefficients, anti-HIV activity, and screening of the “Chimiothèque Nationale” Database), are discussed. Particular attention is paid to mining reaction databases using Condensed Reaction Graphs approach.