Current Topics in Medicinal Chemistry - Volume 12, Issue 18, 2012

Numerical characterization of molecular structure is a first step in many computational analysis of chemical structure data. These numerical representations, termed descriptors, come in many forms, ranging from simple atom counts and invariants of the molecular graph to distribution of properties, such as charge, across a molecular surface. In this article we first present a broad categorization of descriptors and then describe applications and toolkits that can be employed to evaluate them. We highlight a number of issues surrounding molecular descriptor calculations such as versioning and reproducibility and describe how some toolkits have attempted to address these problems.

This manuscript serves as a review of how the R language has been used in the last decade to address problems related to medicinal chemistry design. This includes the use of the R language for chemoinformatics applications and interfaces, as well as statistical modeling for ADMET and potency endpoints. Additionally, a few examples of R code are provided to demonstrate the ability of this language to make available cutting-edge statistical analysis to the medicinal chemistry design community.

Technological advances in high-throughput screening methods, combinatorial chemistry and the design of virtual libraries have evolved in the pursuit of challenging drug targets. Over the last two decades a vast amount of data has been generated within these fields and as a consequence data mining methods have been developed to extract key pieces of information from these large data pools. Much of this data is now available in the public domain. This has been helpful in the arena of drug discovery for both academic groups and for small to medium sized enterprises which previously would not have had access to such data resources. Commercial data mining software is sometimes prohibitively expensive and the alternate open source data mining software is gaining momentum in both academia and in industrial applications as the costs of research and development continue to rise. KNIME, the Konstanz Information Miner, has emerged as a leader in open source data mining tools. KNIME provides an integrated solution for the data mining requirements across the drug discovery pipeline through a visual assembly of data workflows drawing from an extensive repository of tools. This review will examine KNIME as an open source data mining tool and its applications in drug discovery.

We present the open source components for drug discovery that has been developed and integrated into the graphical workbench Bioclipse. Building on a solid open source cheminformatics core, Bioclipse has advanced functionality for managing and visualizing chemical structures and related information. The features presented here include QSAR/QSPR modeling, various predictive solutions such as decision support for chemical liability assessment, site-ofmetabolism prediction, virtual screening, and knowledge discovery and integration. We demonstrate the utility of the described tools with examples from computational pharmacology, toxicology, and ADME. Bioclipse is used in both academia and industry, and is a good example of open source leading to new solutions for drug discovery.

The Structure-Activity Relationships (SAR) landscape and activity cliffs concepts have their origins in medicinal chemistry and receptor-ligand interactions modelling. While intuitive, the definition of an activity cliff as a “pair of structurally similar compounds with large differences in potency” is commonly recognized as ambiguous. This paper proposes a new and efficient method for identifying activity cliffs and visualization of activity landscapes. The activity cliffs definition could be improved to reflect not the cliff steepness alone, but also the rate of the change of the steepness. The method requires explicitly setting similarity and activity difference thresholds, but provides means to explore multiple thresholds and to visualize in a single map how the thresholds affect the activity cliff identification. The identification of the activity cliffs is addressed by reformulating the problem as a statistical one, by introducing a probabilistic measure, namely, calculating the likelihood of a compound having large activity difference compared to other compounds, while being highly similar to them. The likelihood is effectively a quantification of a SAS Map with defined thresholds. Calculating the likelihood relies on four counts only, and does not require the pairwise matrix storage. This is a significant advantage, especially when processing large datasets. The method generates a list of individual compounds, ranked according to the likelihood of their involvement in the formation of activity cliffs, and goes beyond characterizing cliffs by structure pairs only. The visualisation is implemented by considering the activity plane fixed and analysing the irregularities of the similarity itself. It provides a convenient analogy to a topographic map and may help identifying the most appropriate similarity representation for each specific SAR space. The proposed method has been applied to several datasets, representing different biological activities. Finally, the method is implemented as part of an existing open source Ambit package and could be accessed via an OpenTox API compliant web service and via an interactive application, running within a modern, JavaScript enabled web browser. Combined with the functionalities already offered by the OpenTox framework, like data sharing and remote calculations, it could be a useful tool for exploring chemical landscapes online.

Although the motions of proteins are fundamental for their function, for pragmatic reasons, the consideration of protein elasticity has traditionally been neglected in drug discovery and design. This review details protein motion, its relevance to biomolecular interactions and how it can be sampled using molecular dynamics simulations. Within this context, two major areas of research in structure-based prediction that can benefit from considering protein flexibility, binding site detection and molecular docking, are discussed. Basic classification metrics and statistical analysis techniques, which can facilitate performance analysis, are also reviewed. With hardware and software advances, molecular dynamics in combination with traditional structure-based prediction methods can potentially reduce the time and costs involved in the hit identification pipeline.

Driven by a steady improvement of computational hardware and significant progress in ab initio method development, quantum-mechanical approaches can now be applied to large biochemical systems and drug design. We review the methods implemented in GAMESS, which are suitable to calculate large biochemical systems. An emphasis is put on the fragment molecular orbital method (FMO) and quantum mechanics interfaced with molecular mechanics (QM/MM). The use of FMO in the protein-ligand binding, structure-activity relationship (SAR) studies, fragment- and structure-based drug design (FBDD/SBDD) is discussed in detail.