Vishwesh Venkatraman

Comprehensive Comparison of Ligand-Based Virtual Screening Tools Against the DUD Data set Reveals Limitations of Current 3D Methods

In recent years, many virtual screening (VS) tools have been developed that employ different molecular representations and have different speed and accuracy characteristics. In this paper, we compare ten popular ligand-based VS tools using the publicly available Directory of Useful Decoys (DUD) data set comprising over 100 000 compounds distributed across 40 protein targets. The DUD was developed initially to evaluate docking algorithms, but our results from an operational correlation analysis show that it is also well suited for comparing ligand-based VS tools. Although it is conventional wisdom that 3D molecular shape is an important determinant of biological activity, our results based on permutational significance tests of several commonly used VS metrics show that the 2D fingerprint-based methods generally give better VS performance than the 3D shape-based approaches for surprisingly many of the DUD targets. To help understand this finding, we have analyzed the nature of the scoring functions used and the composition of the DUD data set itself. We propose that to improve the VS performance of current 3D methods, it will be necessary to devise screening queries that can represent multiple possible conformations and which can exploit knowledge of known actives that span multiple scaffold families.

Quantitative structure–property relationship modeling of Grätzel solar cell dyes

With fossil fuel reserves on the decline, there is increasing focus on the design and development of low‐cost organic photovoltaic devices, in particular, dye‐sensitized solar cells (DSSCs). The power conversion efficiency (PCE) of a DSSC is heavily influenced by the chemical structure of the dye. However, as far as we know, no predictive quantitative structure‐property relationship models for DSSCs with PCE as one of the response variables have been reported. Thus, we report for the first time the successful application of comparative molecular field analysis (CoMFA) and vibrational frequency‐based eigenvalue (EVA) descriptors to model molecular structure‐photovoltaic performance relationships for a set of 40 coumarin derivatives. The results show that the models obtained provide statistically robust predictions of important photovoltaic parameters such as PCE, the open‐circuit voltage (VOC), short‐circuit current (JSC) and the peak absorption wavelength λmax  . Some of our findings based on the analysis of the models are in accordance with those reported in the literature. These structure‐property relationships can be applied to the rational structural design and evaluation of new photovoltaic materials.

Evolutionary de novo design of phenothiazine derivatives for dye-sensitized solar cells

Traditional approaches for improving the photovoltaic performance of dye-sensitized solar cells (DSSCs) have mainly relied on judicious molecular design and device level modifications. Such schemes, however, are bound by costly and time-consuming synthesis procedures. In this paper, we demonstrate the efficacy of an alternative approach based on in silico evolutionary de novo design of novel dye structures with improved DSSC power conversion efficiency (PCE) values. Because the PCE, cannot as yet be directly computed from first principles, the evolutionary fitness function utilizes predictive structure–property relationship (QSPR) models calibrated from empirical data. Our design approach is applied to phenothiazine-based dye sensitizers. The chemical structure space is explored using a genetic algorithm that systematically assembles molecules from fragments in a synthetically tractable manner. Five novel phenothiazine dyes are proposed using our approach where all have predicted PCE values above 9%.

Automated design of realistic organometallic molecules from fragments.

A method for the automated generation of realistic, synthetically accessible transition metal and organometallic complexes is described. Computational tools were designed to generate molecular fragments, preferably harvested from libraries of existing, stable compounds, to be used as building blocks for the construction of new molecules. These fragments are enriched with information about the number and type of possible connections to other fragments and are stored in library files. When connecting fragments in the subsequent building process, compatibility matrices, which define the connection rules between fragments, are used to delineate organometallic fragment spaces from which molecules can be generated in an automated fashion. The approach is flexible and allows ample structural variation at the same time as the combination of known fragments is easily restrained to avoid generation of exotic and unrealistic substructures and molecules. The method was tested in the generation of ruthenium complexes, with a given coordination environment, which can serve as candidates in catalyst development. The results demonstrate that molecules generated with the described method do not contain exotic arrangements of atoms and are by far more realistic than those obtained by the application of valence rules alone.

Automated Building of Organometallic Complexes from 3D Fragments

A method for the automated construction of three-dimensional (3D) molecular models of organometallic species in design studies is described. Molecular structure fragments derived from crystallographic structures and accurate molecular-level calculations are used as 3D building blocks in the construction of multiple molecular models of analogous compounds. The method allows for precise control of stereochemistry and geometrical features that may otherwise be very challenging, or even impossible, to achieve with commonly available generators of 3D chemical structures. The new method was tested in the construction of three sets of active or metastable organometallic species of catalytic reactions in the homogeneous phase. The performance of the method was compared with those of commonly available methods for automated generation of 3D models, demonstrating higher accuracy of the prepared 3D models in general, and, in particular, a much wider range with respect to the kind of chemical structures that can be built automatically, with capabilities far beyond standard organic and main-group chemistry.

Exploring c-Met kinase flexibility by sampling and clustering its conformational space

It is now widely recognized that the flexibility of both partners has to be considered in molecular docking studies. However, the question how to handle the best the huge computational complexity of exploring the protein binding site landscape is still a matter of debate. Here we investigate the flexibility of c‐Met kinase as a test case for comparing several simulation methods. The c‐Met kinase catalytic site is an interesting target for anticancer drug design. In particular, it harbors an unusual plasticity compared with other kinases ATP binding sites. Exploiting this feature may eventually lead to the discovery of new anticancer agents with exquisite specificity. We present in this article an extensive investigation of c‐Met kinase conformational space using large‐scale computational simulations in order to extend the knowledge already gathered from available X‐ray structures. In the process, we compare the relevance of different strategies for modeling and injecting receptor flexibility information into early stage in silico structure‐based drug discovery pipeline. The results presented here are currently being exploited in on‐going virtual screening investigations on c‐Met. Proteins 2012;.

KRAKENX: software for the generation of alignment-independent 3D descriptors

The KRAKENX software calculates a large variety of molecular descriptors based on quantum chemistry computations. The program supports over 2000 three-dimensional descriptors that are calculated from the output of different quantum chemistry packages. The current implementation supports semi-empirical MOPAC-based computations and primarily focuses on orientation-independent descriptors that have been discussed in the literature. The descriptor performance has been exemplified using a number of large and diverse datasets and can be seen to produce parsimonious linear models. The software can be run on multiple platforms and is available to academics free of charge.

Predicting Multi-Component Protein Assemblies Using an Ant Colony Approach

Biological processes are often governed by functional modules of large protein assemblies such as the proteasomes and the nuclear pore complex, for example. However, atomic structures can be determined experimentally only for a small fraction of these multicomponent assemblies. In this article, we present an ant colony optimization based approach to predict the structure of large multicomponent complexes. Starting with pair-wise docking predictions, a multigraph consisting of vertices representing the component proteins and edges representing scored transformations is constructed. Thus, the assembly problem corresponds to identifying minimum weighted spanning trees that yield arrangements of components with few atomic clashes. The utility of the approach is demonstrated using protein complexes taken from the Protein Data Bank. Our algorithm was able to identify near-native solutions for 5 of the 6 cases tested, including one 6-component complex. This demonstrates that the ant colony model provides a useful way to deal with highly combinatorial problems such as assembling multicomponent protein complexes.

Local initiation conditions for water autoionization

Significance The dissociation of water is arguably the most fundamental chemical reaction occurring in the aqueous phase. Despite that the splitting of a water molecule very seldom occurs, the reaction is of major importance in many areas of chemistry and biology. Direct experimental probing of the event is still impossible and also simulating the event via accurate computer simulations is challenging. Here, we achieved the latter via specialized rare-event algorithms estimating rates of dissociation in agreement with indirect experimental measurements. Even more interestingly, by a rigorous analysis of our results we identified anomalies in the water structure that act as initiators of the reaction, a finding that suggests paradigms for steering and catalyzing chemical reactions. The pH of liquid water is determined by the infrequent process in which water molecules split into short-lived hydroxide and hydronium ions. This reaction is difficult to probe experimentally and challenging to simulate. One of the open questions is whether the local water structure around a slightly stretched OH bond is actually initiating the eventual breakage of this bond or whether this event is driven by a global ordering that involves many water molecules far away from the reaction center. Here, we investigated the self-ionization of water at room temperature by rare-event ab initio molecular dynamics and obtained autoionization rates and activation energies in good agreement with experiments. Based on the analysis of thousands of molecular trajectories, we identified a couple of local order parameters and show that if a bond stretch occurs when all these parameters are around their ideal range, the chance for the first dissociation step (double-proton jump) increases from 10−7 to 0.4. Understanding these initiation triggers might ultimately allow the steering of chemical reactions.

Benzotriazoles, benzothiazoles and trace elements in an urban road setting in Trondheim, Norway: Re-visiting the chemical markers of traffic pollution

Road traffic emissions are known to contribute heavily to the pollution in urban environments. The aim of this study was to establish specific traffic pollution markers in an urban road setting based on the occurrence profiles of benzotriazoles, benzothiazoles and trace elements in road dust and relevant matrices, including airborne particulate matter and core asphalt. Benzotriazoles and benzothiazoles are high-production volume chemicals that are used as complexing and anticorrosive agents for metals, act as vulcanizing accelerators for rubber materials, and possess anti-freezing/anti-icing properties. In this study, six benzothiazoles (benzothiazole, 2‑morpholin‑4‑yl‑benzothiazole, 2‑hydroxy‑benzothiazole, 2‑thio‑benzothiazole, 2‑methylthio‑benzothiazole, and 2‑amino‑benzothiazole), seven benzotriazoles (1H‑benzotriazole, 1‑hydroxy‑benzotriazole, 5‑chloro‑1H‑benzotriazole, tolyltriazole, xylyltriazole, benzotriazole‑5‑carboxyl acid, and 5‑amino‑1H‑benzotriazole), and 66 trace elements were determined in road dust samples from a sub-arctic urban road setting in Norway, and seasonal occurrence profiles were assessed between the studded and the non-studded tire season. The road dust was collected as suspended particulate matter in an aqueous phase with the introduced dust sampler in Scandinavia, the Wet Dust Sampler. The concentrations of the sum of seven benzotriazoles (Σ(7)BTRs) and six benzothiazoles (Σ(6)BTHs) in road dust ranged from 191 to 3054 ng/L and 93.4 to 1903 ng/L, respectively. To the best of our knowledge, 1H‑benzotriazole and tolyltriazole are reported for the first time as suitable markers of metal corrosion in vehicles. From the benzothiazole class, 2‑thio‑benzothiazole was found to be a suitable marker of tire rubber particles, while its methylated derivative, 2‑methylthio‑benzothiazole, was found to be a marker of chemical leaching. In addition, different types of new unused tires (summer, studded, and non-studded) were analyzed to assess their benzothiazoles and benzotriazoles content. Based on the concentrations found for benzotriazoles and benzothiazoles in airborne particulate matter, human exposure doses were calculated, and the estimated daily intake doses were found on the order of picograms per day.