The beauty of molecular surfaces as revealed by self-organizing neural networks

Computational insight into the chemical space of plant growth regulators

An enormous technological progress has resulted in an explosive growth in the amount of biological and chemical data that is typically multivariate and tangled in structure. Therefore, several computational approaches have mainly focused on dimensionality reduction and convenient representation of high-dimensional datasets to elucidate the relationships between the observed activity (or effect) and calculated parameters commonly expressed in terms of molecular descriptors. We have collected the experimental data available in patent and scientific publications as well as specific databases for various agrochemicals. The resulting dataset was then thoroughly analyzed using Kohonen-based self-organizing technique. The overall aim of the presented study is to investigate whether the developed in silico model can be applied to predict the agrochemical activity of small molecule compounds and, at the same time, to offer further insights into the distinctive features of different agrochemical categories. The preliminary external validation with several plant growth regulators demonstrated a relatively high prediction power (67%) of the constructed model. This study is, actually, the first example of a large-scale modeling in the field of agrochemistry.

Deciphering fine molecular details of proteins' structure and function with a Protein Surface Topography (PST) method

Molecular surfaces are the key players in biomolecular recognition and interactions. Nowadays, it is trivial to visualize a molecular surface and surface-distributed properties in three-dimensional space. However, such a representation trends to be biased and ambiguous in case of thorough analysis. We present a new method to create 2D spherical projection maps of entire protein surfaces and manipulate with them--protein surface topography (PST). It permits visualization and thoughtful analysis of surface properties. PST helps to easily portray conformational transitions, analyze proteins' properties and their dynamic behavior, improve docking performance, and reveal common patterns and dissimilarities in molecular surfaces of related bioactive peptides. This paper describes basic usage of PST with an example of small G-proteins conformational transitions, mapping of caspase-1 intersubunit interface, and intrinsic "complementarity" in the conotoxin-acetylcholine binding protein complex. We suggest that PST is a beneficial approach for structure-function studies of bioactive peptides and small proteins.

Fractal nature of protein surface roughness: a note on quantification of change of surface roughness in active sites, before and after binding

Year 2010 marked the 25th year since we came to know that roughness of a protein surface has fractal symmetry. Ever since the publication of Lewis and Rees' paper, hundreds of works from a spectrum of perspectives have established that fractal dimension (FD) can be considered as a reliable marker that describes roughness of protein surface objectively. In this article, we introduce readers to the fundamentals of fractals and present categorical biophysical and geometrical reasons as to why FD-based constructs can describe protein surface roughness more accurately. We then review the commonality (and the lack of it) between numerous approaches that have attempted to investigate protein surface with fractal measures, before exploring the patterns in the results that they have produced. Apart from presenting the genealogy of approaches and results, we present an analysis that quantifies the difference in surface roughness in stretches of protein surface containing the active site, before and after binding to ligands, to underline the utility of FD-based measures further. It has been found that surface stretches containing the active site, in general, undergo a significant increment in its roughness after binding. After presenting the entire repertoire of FD-based surface roughness studies, we talk about two yet-unexplored problems where application of FD-based techniques can help in deciphering underlying patterns of surface interactions. Finally, we list the limitations of FD-based constructs and put down several precautions that one must take while working with them.

Self-Organizing Maps for In Silico Screening and Data Visualization

Self-organizing maps, which are unsupervised artificial neural networks, have become a very useful tool in a wide area of disciplines, including medicinal chemistry. Here, we will focus on two applications of self-organizing maps: the use of self-organizing maps for in silico screening and for clustering and visualisation of large datasets. Additionally, the importance of parameter selection is discussed and some modifications to the original algorithm are summarised.

Network models in drug discovery and regenerative medicine

Network motifs and modelling paradigms are attracting increasing attention as modelling tools in drug design and development, and in regenerative medicine. There is a gradual but inexorable convergence between these hitherto disparate disciplines. This review summarizes some very recent work in these areas, leading to an understanding of the complementary roles networks play and factors driving this convergence: network paradigms can be excellent ways of modelling and understanding drug molecules and their action, an understanding of the robustness and vulnerabilities of biological targets may improve the efficacy of drug design and discovery, drug design has an increasingly large role to play in directing stem cell properties, stem cell regulatory networks can be modelled in useful ways using network models at a reasonable level of scale, and the network tools of drug design are also very useful for the design of biomaterials used in regenerative medicine.

The N-terminal domain of 2',3'-cyclic nucleotide 3'-phosphodiesterase harbors a GTP/ATP binding site

The interaction between 2',3'-cyclic nucleotide 3'-phosphodiesterase and guanine/adenine nucleotides was investigated. The binding of purine nucleotides to 2',3'-cyclic nucleotide 3'-phosphodiesterase was revealed by both direct and indirect methods. In fact, surface plasmon resonance experiments, triphosphatase activity measurements, and fluorescence experiments revealed that 2',3'-cyclic nucleotide 3'-phosphodiesterase binds purine nucleotide triphosphates with an affinity higher than that displayed for diphosphates; on the contrary, the affinity for both purine monophosphates and pyrimidine nucleotides was negligible. An interpretation of biological experimental data was achieved by a building of 2',3'-cyclic nucleotide 3'-phosphodiesterase N-terminal molecular model. The structural elements responsible for nucleotide binding were identified and potential complexes between the N-terminal domain of CNP-ase and nucleotide were analyzed by docking simulations. Therefore, our findings suggest new functional and structural property of the N-terminal domain of CNPase.

SOMMER: self-organising maps for education and research

SOMMER is a publicly available, Java-based toolbox for training and visualizing two- and three-dimensional unsupervised self-organizing maps (SOMs). Various map topologies are implemented for planar rectangular, toroidal, cubic-surface and spherical projections. The software allows for visualization of the training process, which has been shown to be particularly valuable for teaching purposes.

Self-organizing neural networks for pharmacophore mapping

We have shown that the SOM network can be a useful tool in pharmacophore mapping strategy. A possibility for the generation of fuzzy molecular representations together with its ability for discovering such aspects of molecular similarity that can be easily overlooked by a human chemist is an important advantage. The reduction in complexity resulting from the data compression is another one. The main disadvantage of SOM usage is the need for the application of special software packages not usually organized in user friendly toolboxes that can be applied easily. Instead, it needs some experience and time to optimize the parameters controlling the performance of the network.

Artificial neural network prediction of aerosol deposition in human lungs

PURPOSE

To develop a rapid and reliable method for predicting the pattern of aerosol particle deposition within the human lungs, using artificial neural networks (ANNs).

METHODS

Experimental data from the literature were used to train multi-layer perceptron (MLP) networks to allow for prediction of regional and total aerosol particle deposition patterns in human lungs. These data covered particle sizes in the range 0.05-15 microm and three different breathing patterns (ranging from "quiet" breathing to breathing "under physical work conditions"). Three different MLPs were trained, to provide separate predictions of aerosol particle deposition in the laryngeal, bronchial, and alveolar regions. The total deposition fraction for a given set of breathing conditions was computed simply as the sum of the outputs produced from the corresponding regional deposition MLPs.

RESULTS

The ANNs developed are shown to give highly accurate predictions for both regional and total aerosol deposition patterns for all particle sizes and breathing conditions (with errors typically less than 0.04%).

CONCLUSIONS

We conclude that the current set of ANNs can be used to give good predictions of particle deposition from polydisperse pharmaceutical aerosols generated from breath-actuated dry powder inhalers, nebulizers, and metered dose inhalers with spacers.

An integrated SOM-fuzzy ARTMAP neural system for the evaluation of toxicity

Self-organized maps (SOM) have been applied to analyze the similarities of chemical compounds and to select from a given pool of descriptors the smallest and more relevant subset needed to build robust QSAR models based on fuzzy ARTMAP. First, the category maps for each molecular descriptor and for the target activity variable were created with SOM and then classified on the basis of topology and nonlinear distribution. The best subset of descriptors was obtained by choosing from each cluster the index with the highest correlation with the target variable and then in order of decreasing correlation. This process was terminated when a dissimilarity measure increased, indicating that the inclusion of more molecular indices would not add supplementary information. The optimal subset of descriptors was used as input to a fuzzy ARTMAP architecture modified to effect predictive capabilities. The performance of the integrated SOM-fuzzy ARTMAP approach was evaluated with the prediction of the acute toxicity LC50 of a homogeneous set of 69 benzene derivatives in the fathead minnow and the oral rat toxicity LD50 of a heterogeneous set of 155 organic compounds. The proposed methodology minimized the problem of misclassification of similar compounds and significantly enhanced the predictive capabilities of a properly trained fuzzy ARTMAP network.

Steroid binding by antibodies and artificial receptors: exploration of theoretical methods to determine the origins of binding affinities and specificities

Binding mode calculations for complexes between an artificial paracyclophane receptor and digoxins, cholic acids as well as cortisone steroids show encapsulation of different ring combinations. Docking experiments were performed between the 26-10 antibody and digoxins. Coordination affinity arises from hydrophobic desolvation and van der Waals interactions rather than from hydrogen bonds. The specificity and affinity arises mainly from shape complementarity. Computed binding free energies and Kohonen neural network computations both point to physicochemical and structural similarities of natural antibodies and artificial receptors.

Application of Kohonen Neural Networks in classification of biologically active compounds

Automated data classification is an indispensable tool in Drug Design. It allows to select homogeneous training sets or to distinguish compounds with required biological properties. The Kohonen Neural Networks (KNN) suggest new means for classification of biologically interesting compounds. In this paper, first, capabilities of KNN in data dimensionality reduction are presented as compared with the capabilities of Principal Component Analysis (PCA) and Hierarchical Cluster Analysis (HCA). The advantages of KNN become evident with increasing data dimensionality and size of the training set. Then, new methods are suggested to evaluate the quality of KNN models. Finally, a case study on chemical and biological data is presented. The database studied includes more than 2000 organophosphorous potent pesticides. The Kohonen maps were obtained which allow to distinguish compounds with different biological behavior.

Analytically defined surfaces to analyze molecular interaction properties

Molecular surfaces are widely used for characterizing molecules and displaying and quantifying their interaction properties. Here we consider molecular surfaces defined as isocontours of a function (a sum of exponential functions centered on each atom) that approximately represents electron density. The smoothness is advantageous for surface mapping of molecular properties (e.g., electrostatic potential). By varying parameters, these surfaces can be constructed to represent the van der Waals or solvent-accessible surface of a molecular with any accuracy. We describe numerical algorithms to operate on the analytically defined surfaces. Two applications are considered: (1) We define and locate extremal points of molecular properties on the surfaces. The extremal points provide a compact representation of a property on a surface, obviating the necessity to compute values of the property on an array of surface points as is usually done; (2) a molecular surface patch or interface is projected onto a flat surface (by introducing curvilinear coordinates) with approximate conservation of area for analysis purposes. Applications to studies of protein-protein interactions are described.

The comparison of geometric and electronic properties of molecular surfaces by neural networks: application to the analysis of corticosteroid-binding globulin activity of steroids

It is shown how a self-organizing neural network such as the one introduced by Kohonen can be used to analyze features of molecular surfaces, such as shape and the molecular electrostatic potential. On the one hand, two-dimensional maps of molecular surface properties can be generated and used for the comparison of a set of molecules. On the other hand, the surface geometry of one molecule can be stored in a network and this network can be used as a template for the analysis of the shape of various other molecules. The application of these techniques to a series of steroids exhibiting a range of binding activities to the corticosteroid-binding globulin receptor allows one to pinpoint the essential features necessary for biological activity.

Locating biologically active compounds in medium-sized heterogeneous datasets by topological autocorrelation vectors: dopamine and benzodiazepine agonists

Electronic properties located on the atoms of a molecule such as partial atomic charges as well as electronegativity and polarizability values are encoded by an autocorrelation vector accounting for the constitution of a molecule. This encoding procedure is able to distinguish between compounds being dopamine agonists and those being benzodiazepine receptor agonists even after projection into a two-dimensional self-organizing network. The two types of compounds can still be distinguished if they are buried in a dataset of 8323 compounds of a chemical supplier catalog comprising a wide structural variety. The maps obtained by this sequence of events, calculation of empirical physicochemical effects, encoding in a topological autocorrelation vector, and projection by a self-organizing neural network, can thus be used for searching for structural similarity, and, in particular, for finding new lead structures with biological activity.

Comparison of structurally different allosteric modulators of muscarinic receptors by self-organizing neural networks

Similarities in the molecular structure and surface properties of the allosteric modulators of muscarinic receptors, alcuronium, gallamine, tubocurarine, and the hexamethonium compound W84, a well-known pharmacological tool, are explored. The analysis of the molecular electrostatic potential (MEP) as well as of the shape of the molecular surface is performed by self-organizing neural networks. A distorted sandwich conformation of W84 is suggested to be the active form. The importance of the MEP for binding of these compounds could be established.

Neural network computer simulation of medical aerosols

Preliminary investigations have been conducted to assess the potential for using artificial neural networks to simulate aerosol behaviour, with a view to employing this type of methodology in the evaluation and design of pulmonary drug-delivery systems. Details are presented of the general purpose software developed for these tasks; it implements a feed-forward back-propagation algorithm with weight decay and connection pruning, the user having complete run-time control of the network architecture and mode of training. A series of exploratory investigations is then reported in which different network structures and training strategies are assessed in terms of their ability to simulate known patterns of fluid flow in simple model systems. The first of these involves simulations of cellular automata-generated data for fluid flow through a partially obstructed two-dimensional pipe. The artificial neural networks are shown to be highly successful in simulating the behaviour of this simple linear system, but with important provisos relating to the information content of the training data and the criteria used to judge when the network is properly trained. A second set of investigations is then reported in which similar networks are used to simulate patterns of fluid flow through aerosol generation devices, using training data furnished through rigorous computational fluid dynamics modelling. These more complex three-dimensional systems are modelled with equal success. It is concluded that carefully tailored, well trained networks could provide valuable tools not just for predicting but also for analysing the spatial dynamics of pharmaceutical aerosols.

New approaches in molecular structure prediction

In the past years, much effort has been put on the development of new methodologies and algorithms for the prediction of protein secondary and tertiary structures from (sequence) data; this is reviewed in detail. New approaches for these predictions such as neural network methods, genetic algorithms, machine learning, and graph theoretical methods are discussed. Secondary structure prediction algorithms were improved mostly by considering families of related proteins; however, for the reliable tertiary structure modeling of proteins, knowledge-based techniques are still preferred. Methods and examples with more or less successful results are described. Also, programs and parameterizations for energy minimisations, molecular dynamics, and electrostatic interactions have been improved, especially with respect to their former limits of applicability. Other topics discussed in this review include the use of traditional and on-line databases, the docking problem and surface properties of biomolecules, packing of protein cores, de novo design and protein engineering, prediction of membrane protein structures, the verification and reliability of model structures, and progress made with currently available software and computer hardware. In summary, the prediction of the structure, function, and other properties of a protein is still possible only within limits, but these limits continue to be moved.