Lazaros Mavridis

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 100u2009000 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.

Toward high throughput 3D virtual screening using spherical harmonic surface representations

Searching chemical databases for possible drug leads is often one of the main activities conducted during the early stages of a drug development project. This article shows that spherical harmonic molecular shape representations provide a powerful way to search and cluster small-molecule databases rapidly and accurately. Our clustering results show that chemically meaningful clusters may be obtained using only low order spherical harmonic expansions. Our database search results show that using low order spherical harmonic shape-based correlation techniques could provide a practical and efficient way to search very large 3D molecular databases, hence leading to a useful new approach for high throughput 3D virtual screening. The approach described is currently being extended to allow the rapid search and comparison of arbitrary combinations of molecular surface properties.

3D-blast: 3D protein structure alignment, comparison, and classification using spherical polar Fourier correlations.

This paper presents a novel sequence-independent method of aligning protein structures using three-dimensional spherical polar Fourier (SPF) representations of protein shape. The approach is demonstrated by clustering subsets of the CATH database for each of the four main CATH fold types, and by searching the entire CATH database of some 12,000 structures using several protein structures as queries. Overall, the automatic SPF clustering approach agrees very well with the expert-curated CATH classification, and ROC plot analyses of the database searches show that the approach has very high precision and recall. Database query times can be reduced considerably by using a simple rotationally-invariant pre-filter in tandem with a more sensitive rotational search with little or no reduction in accuracy. Hence it should soon be possible to perform on-line 3D structural searches in interactive time-scales.

Detecting drug promiscuity using Gaussian ensemble screening.

Polypharmacology describes the binding of a ligand to multiple protein targets (a promiscuous ligand) or multiple diverse ligands binding to a given target (a promiscuous target). Pharmaceutical companies are discovering increasing numbers of both promiscuous drugs and drug targets. Hence, polypharmacology is now recognized as an important aspect of drug design. Here, we describe a new and fast way to predict polypharmacological relationships between drug classes quantitatively, which we call Gaussian Ensemble Screening (GES). This approach represents a cluster of molecules with similar spherical harmonic surface shapes as a Gaussian distribution with respect to a selected center molecule. Calculating the Gaussian overlap between pairs of such clusters allows the similarity between drug classes to be calculated analytically without requiring thousands of bootstrap comparisons, as in current promiscuity prediction approaches. We find that such cluster similarity scores also follow a Gaussian distribution. Hence, a cluster similarity score may be transformed into a probability value, or p-value, in order to quantify the relationships between drug classes. We present results obtained when using the GES approach to predict relationships between drug classes in a subset of the MDL Drug Data Report (MDDR) database. Our results indicate that GES is a useful way to study polypharmacology relationships, and it could provide a novel way to propose new targets for drug repositioning.

Recent trends and applications in 3D virtual screening.

Virtual screening (VS) is becoming an increasingly important approach for identifying and selecting biologically active molecules against specific pharmaceutically relevant targets. Compared to conventional high throughput screening techniques, in silico screening is fast and inexpensive, and is increasing in popularity in early-stage drug discovery endeavours. This paper reviews and discusses recent trends and developments in three-dimensional (3D) receptor-based and ligand-based VS methodologies. First, we describe the concept of accessible chemical space and its exploration. We then describe 3D structural ligand-based VS techniques, hybrid approaches, and new approaches to exploit additional knowledge that can now be found in large chemogenomic databases. We also briefly discuss some potential issues relating to pharmacokinetics, toxicity profiling, target identification and validation, inverse docking, scaffold-hopping and drug re-purposing. We propose that the best way to advance the state of the art in 3D VS is to integrate complementary strategies in a single drug discovery pipeline, rather than to focus only on theoretical or computational improvements of individual techniques. Two recent 3D VS case studies concerning the LXR-β receptor and the CCR5/CXCR4 HIV co-receptors are presented as examples which implement some of the complementary methods and strategies that are reviewed here.

Using Spherical Harmonic Surface Property Representations for Ligand-Based Virtual Screening.

Ligand‐based virtual screening (VS) techniques have become well established in the drug discovery process. However, despite their relative success, there still exists the problem of how to define the initial query compounds and which of their conformations should be used. Here, we propose a novel shape plus surface property approach using multiple local spherical harmonic (SH) functions. We also investigate the use of shape‐based and shape plus property‐based consensus SH queries calculated in several different ways. The utility of these approaches is compared using the 40 pharmaceutically relevant targets of the DUD database. Our results show that using a combination of SH‐based properties often gives better VS performance than using simple shape‐based queries. Shape‐based consensus queries also perform well, but we find that explicit 3D shape‐property conformations should be retained for highly flexible ligands.

Predicting drug promiscuity using spherical harmonic surface shape-based similarity comparisons

Polypharmacology is becoming an increasingly important aspect in drug design. Pharmaceutical companies are discovering more and more cases in which multiple drugs bind to a given target (promiscuous targets) and in which a given drug binds to more than one target (promiscuous ligands). These phenomena are clearly of great importance when considering drug side-effects. In the last 4 years, more than 30 drugs have been tested against more than 40 novel secondary targets based on promiscuity predictions. Current methods for predicting promiscuity typically aim to relate protein receptors according to their primary sequences, the similarity of their ligands, and more recently, the similarity of their ligand binding pockets.

SHREC'10 track: protein model classification

This paper presents the results of the 3D Shape Retrieval Contest 2010 (SHREC10) track Protein Models Classification. The aim of this track is to evaluate how well 3D shape recognition algorithms can classify protein structures according to the CATH [CSL?08] superfamily classification. Five groups participated in this track, using a total of six methods, and for each method a set of ranked predictions was submitted for each classification task. The evaluation of each method is based on the nearest neighbour and area under the curve(AUC) metrics.

Representing and comparing protein folds and fold families using three‐dimensional shape‐density representations

The question of how best to compare and classify the (three‐dimensional) structures of proteins is one of the most important unsolved problems in computational biology. To help tackle this problem, we have developed a novel shape‐density superposition algorithm called 3D‐Blast which represents and superposes the shapes of protein backbone folds using the spherical polar Fourier correlation technique originally developed by us for protein docking. The utility of this approach is compared with several well‐known protein structure alignment algorithms using receiver‐operator‐characteristic plots of queries against the “gold standard” CATH database. Despite being completely independent of protein sequences and using no information about the internal geometry of proteins, our results from searching the CATH database show that 3D‐Blast is highly competitive compared to current state‐of‐the‐art protein structure alignment algorithms. A novel and potentially very useful feature of our approach is that it allows an average or “consensus” fold to be calculated easily for a given group of protein structures. We find that using consensus shapes to represent entire fold families also gives very good database query performance. We propose that using the notion of consensus fold shapes could provide a powerful new way to index existing protein structure databases, and that it offers an objective way to cluster and classify all of the currently known folds in the protein universe. Proteins 2012.

Predicting drug polypharmacology using a novel surface property similarity-based approach

In recent years, polypharmacology is becoming anincreasingly important aspect in drug design. For exam-ple, pharmaceutical companies are discovering more andmore cases in which multiple drugs bind to a given tar-get (promiscuous targets) and in which a given drugbinds to more than one target (promiscuous ligands).Both of these phenomena are clearly of great impor-tance when considering drug side-effects. Given thatscreening drugs against all the proteins expressed by thehuman genome is infeasible, several computational tech-niques for predicting the pharmacological profiles ofdrugs have been developed, ranging from statistical ana-lyses of chemical fingerprints and biological activities [1]to 3D docking of ligand structures into protein pockets.Here we present a novel shape-based approach whichuses spherical harmonic (SH) representations [2,3] tocompare molecular surfaces and key surface propertiesvery efficiently. This approach compares targets by theSH similarity of their ligands and also of their bindingpockets. This allows promiscuous ligands and targets tobe identified and characterized.In this contribution, we present details of ourapproach applied to a subset of the MDL Drug DataReport (MDDR) database containing 65367 compoundsdistributed over 249 diverse pharmacological targets forwhich experimental binding information is known. Thesimilarity of each ligand to each target’s ligand set isquantified and used to predict promiscuity. To ourknowledge, this is the largest all-against-all polypharma-cological study to have been carried out using shape-based techniques. We compare our promiscuity predic-tions with computational and experimental resultsobtained by Keiser et al.[4]. We also analyse thecorrelation between binding pocket shapes and ligand-based promiscuity predictions using the ligand andpocket shape similarity matrices.