Multiple flexible alignment with SEAL: a study of molecules acting on the colchicine binding site

An extension of the steric and electrostatic alignment alignment (SEAL) method (MultiSEAL) is described that allows the overlay of multiple molecules and conformations. The method is well-suited for the systematic study of possible alignments, also revealing information about the conformational energies associated with a given overlay. It has been tested on three examples: angiotensin II antagonists, 5-HT3 antagonists, and dopaminergic compounds. The utility of the method is further demonstrated in an analysis of molecules that putatively bind to the colchicine site of tubulin. On the basis of its overlay with colchicine, allocolchicine, 2-methoxy-5-(2',3',4'-trimethoxyphenyl)tropone, and combretastatin A-4, it appears that 2-methoxyestradiol (2-ME) is unlikely to fit the colchine site properly. The weak antimitotic activity of 2-ME may be explained by its partial fit in the site.     

Alignment of molecules by the Monte Carlo optimization of molecular similarity indices

3D-QSAR uses statistical techniques to correlate calculated structural properties with target properties like biological activity. The comparison of calculated structural properties is dependent upon the relative orientations of molecules in a given data set. Typically molecules are aligned by performing an overlap of common structural units. This “alignment rule” is adequate for a data set, that is closely related structurally, but is far more difficult to apply to either a diverse data set or on the basis of some structural property other than shape, even for sterically similar molecules. In this work we describe a new algorithm for molecular alignment based upon optimization of molecular similarity indices. We show that this Monte Carlo based algorithm is more effective and robust than other optimizers applied previously to the similarity based alignment problem. We show that QSARs derived using the alignments generated by our algorithm are superior to QSARs derived using the more common alignment of fitting of common structural units. © 1997 by John Wiley & Sons, Inc. J Comput Chem 18 : 1344–1353, 1997     

FLAME: a program to flexibly align molecules

Herein, we describe a method to flexibly align molecules (FLAME = FLexibly Align MolEcules). FLAME aligns two molecules by first finding maximum common pharmacophores between them using a genetic algorithm. The resulting alignments are then subjected to simultaneous optimizations of their internal energies and an alignment score. The utility of the method in pairwise alignment, multiple molecule flexible alignment, and database searching was examined. For pairwise alignment, two carboxypeptidase ligands (Protein Data Bank codes and ), two estrogen receptor ligands ( and ), and two thrombin ligands ( and ) were used as test sets. Alignments generated by FLAME starting from CONCORD structures compared very well to the X-ray structures (average root-mean-square deviation = 0.36 A) even without further minimization in the presence of the protein. For multiple flexible alignments, five structurally diverse D3 receptor ligands were used as a test set. The FLAME alignment automatically identified three common pharmacophores: a base, a hydrogen-bond acceptor, and a hydrophobe/aromatic ring. The best alignment was then used to search the MDDR database. The search results were compared to the results using atom pair and Daylight fingerprint similarity. A similar database search comparison was also performed using estrogen receptor modulators. In both cases, hits identified by FLAME were structurally more diverse compared to those from the atom pair and Daylight fingerprint methods.     

Toward a global maximization of the molecular similarity function: Superposition of two molecules

A quantum similarity measure between two molecules is normally identified with the maximum value of the overlap of the corresponding molecular electron densities. The electron density overlap is a function of the mutual positioning of the compared molecules, requiring the measurement of similarity, a solution of a multiple-maxima problem. Collapsing the molecular electron densities into the nuclei provides the essential information toward a global maximization of the overlap similarity function, the maximization of which, in this limit case, appears to be related to the so-called assignment problem. Three levels of approach are then proposed for a global search scanning of the similarity function. In addition, atom—atom similarity Lorentzian potential functions are defined for a rapid completion of the function scanning. Performance is tested among these three levels of simplification and the Monte Carlo and simplex methods. Results reveal the present algorithms as accurate, rapid, and unbiased techniques for density-based molecular alignments. © 1997 by John Wiley & Sons, Inc. J Comput Chem 18: 826–846, 1997     

MARS: Computing Three-Dimensional Alignments for Multiple Ligands Using Pairwise Similarities

The three-dimensional (3D) superimposition of molecules of one biological target reflecting their relative bioactive orientation is key for several ligand-based drug design studies (e.g., QSAR studies, pharmacophore modeling). However, with the lack of sufficient ligand-protein complex structures, an experimental alignment is difficult or often impossible to obtain. Several computational 3D alignment tools have been developed by academic or commercial groups to address this challenge. Here, we present a new approach, MARS (Multiple Alignments by ROCS-based Similarity), that is based on the pairwise alignment of all molecules within the data set using the tool ROCS (Rapid Overlay of Chemical Structures). Each pairwise alignment is scored, and the results are captured in a score matrix. The ideal superimposition of the compounds in the set is then identified by the analysis of the score matrix building stepwise a superimposition of all molecules. The algorithm exploits similarities among all molecules in the data set to compute an optimal 3D alignment. This alignment tool presented here can be used for several applications, including pharmacophore model generation, 3D QSAR modeling, 3D clustering, identification of structural outliers, and addition of compounds to an already existing alignment. Case studies are shown, validating the 3D alignments for six different data sets.     

Relative binding orientations of adenosine A1 receptor ligands — A test case for Distributed Multipole Analysis in medicinal chemistry

Summary The electrostatic properties of adenosine-based agonists and xanthine-based antagonists for the adenosine A₁ receptor were used to assess various proposals for their relative orientation in the unknown binding site. The electrostatic properties were calculated from distributed multipole representations of SCF wavefunctions. A range of methods of assessing the electrostatic similarity of the ligands were used in the comparison. One of the methods, comparing the sign of the potential around the two molecules, gave inconclusive results. The other approaches, however, provided a mutually complementary and consistent picture of the electrostatic similarity and dissimilarity of the molecules in the three proposed relative orientations. This was significantly different from the results obtained previously with MOPAC AM1 point charges. In the standard model overlay, where the aromatic nitrogen atoms of both agonists and antagonists are in the same position relative to the binding site, the electrostatic potentials are so dissimilar that binding to the same receptor site is highly unlikely. Overlaying the N⁶-region of adenosine with that near C8 of theophylline (the N⁶-C8 model) produces the greatest similarity in electrostatic properties for these ligands. However, N⁶-cyclopentyladenosine (CPA) and 1,3-dipropyl-8-cyclopentylxanthine (DPCPX) show greater electrostatic similarity when the aromatic rings are superimposed according to the flipped model, in which the xanthine ring is rotated around its horizontal axis. This difference is mainly attributed to the change in conformation of N⁶-substituted adenosines and could result in a different orientation for theophylline and DPCPX within the receptor binding site. However, it is more likely that DPCPX also binds according to the N⁶-C8 model, as this model gives the best steric overlay and would be favoured by the lipophilic forces, provided that the binding site residues could accommodate the different electrostatic properties in the N⁶/N7-region. Finally, we have shown that Distributed Multipole Analysis (DMA) offers a new, feasible tool for the medicinal chemist, because it provides the use of reliable electrostatic models to determine plausible relative binding orientations.Abbreviations DMA distributed multipole analysis - SCF self-consistent field - CPA N⁶-cyclopentyladenosine - DPCPX 1,3-dipropyl-8-cyclopentylxanthine - R-PIA R-1-phenyl-2-propyladenosine - S-PIA S-1-phenyl-2-propyladenosine     

Efficient overlay of small organic molecules using 3D pharmacophores

Aligning and overlaying two or more bio-active molecules is one of the key tasks in computational drug discovery and bio-activity prediction. Especially chemical-functional molecule characteristics from the view point of a macromolecular target represented as a 3D pharmacophore are the most interesting similarity measure when describing and analyzing macromolecule-ligand interaction. In this study, a novel approach for aligning rigid three-dimensional molecules according to their chemical-functional pharmacophoric features is presented and compared to the overlay of experimentally determined poses in a comparable macromolecule coordinate frame. The presented approach identifies optimal chemical feature pairs using distance and density characteristics obtained by correlating pharmacophoric geometries and thus proves to be faster than existing combinatorial alignment methods and creates more reasonable alignments than pure atom-based methods. Examples will be provided to demonstrate the feasibility, speed and intuitiveness of this method.     

Testing homology with Contact Accepted mutatiOn (CAO): a contact-based Markov model of protein evolution

Point Accepted Mutation (PAM) is the Markov model of amino acid replacements in proteins introduced by Dayhoff and her co-workers (Dayhoff et al., 1978). The PAM matrices and other matrices based on the PAM model have been widely accepted as the standard scoring system of protein sequence similarity in protein sequence alignment tools. Here, we present Contact Accepted mutatiOn (CAO), a Markov model of protein residue contact mutations. The CAO model simulates the interchanging of structurally defined side-chain contacts, and introduces additional structural information into protein sequence alignments. Therefore, similarities between structurally conserved sequences can be detected even without apparent sequence similarity. CAO has been benchmarked on the HOMSTRAD database and a subset of the CATH database, by comparing sequence alignments with reference alignments derived from structural superposition. CAO yields scores that reflect coherently the structural quality of sequence alignments, which has implications particularly for homology modelling and threading techniques.     

4D-QSAR analysis of a series of antifungal p450 inhibitors and 3D-pharmacophore comparisons as a function of alignment

A training set of 55 antifungal p450 analogue inhibitors was used to construct receptor-independent four-dimensional quantitative structure-activity relationship (RI 4D-QSAR) models. Ten different alignments were used to build the models, and one alignment yields a significantly better model than the other alignments. Two different methodologies were used to measure the similarity of the best 4D-QSAR models of each alignment. One method compares the residual of fit between pairs of models using the cross-correlation coefficient of their residuals of fit as a similarity measure. The other method compares the spatial distributions of the IPE types (3D-pharmacophores) of pairs of 4D-QSAR models from different alignments. Optimum models from several different alignments have nearly the same correlation coefficients, r(2), and cross-validation correlation coefficients, xv-r(2), yet the 3D-pharmacophores of these models are very different from one another. The highest 3D-pharmacophore similarity correlation coefficient between any pair of 4D-QSAR models from the 10 alignments considered is only 0.216. However, the best 4D-QSAR models of each alignment do contain some proximate common pharmacorphore sites. A test set of 10 compounds was used to validate the predictivity of the best 4D-QSAR models of each alignment. The "best" model from the 10 alignments has the highest predictivity. The inferred active sites mapped out by the 4D-QSAR models suggest that hydrogen bond interactions are not prevalent when this class of P450 analogue inhibitors binds to the receptor active site. This feature of the 4D-QSAR models is in agreement with the crystal structure results that indicate no ligand-receptor hydrogen bonds are formed.     

3D QSAR studies on protein tyrosine phosphatase 1B inhibitors: comparison of the quality and predictivity among 3D QSAR models obtained from different conformer-based alignments

A set of 65 flexible peptidomimetic competitive inhibitors (52 in the training set and 13 in the test set) of protein tyrosine phosphatase 1B (PTP1B) has been used to compare the quality and predictive power of 3D quantitative structure-activity relationship (QSAR) comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) models for the three most commonly used conformer-based alignments, namely, cocrystallized conformer-based alignment (CCBA), docked conformer-based alignment (DCBA), and global minima energy conformer-based alignment (GMCBA). These three conformers of 5-[(2S)-2-({(2S)-2-[(tert-butoxycarbonyl)amino]-3-phenylpropanoyl}amino)3-oxo-3-pentylamino)propyl]-2-(carboxymethoxy)benzoic acid (compound number 66) were obtained from the X-ray structure of its cocrystallized complex with PTP1B (PDB ID: 1JF7), its docking studies, and its global minima by simulated annealing. Among the 3D QSAR models developed using the above three alignments, the CCBA provided the optimal predictive CoMFA model for the training set with cross-validated r2 (q2)=0.708, non-cross-validated r2=0.902, standard error of estimate (s)=0.165, and F=202.553 and the optimal CoMSIA model with q2=0.440, r2=0.799, s=0.192, and F=117.782. These models also showed the best test set prediction for the 13 compounds with predictive r2 values of 0.706 and 0.683, respectively. Though the QSAR models derived using the other two alignments also produced statistically acceptable models in the order DCBA>GMCBA in terms of the values of q2, r2, and predictive r2, they were inferior to the corresponding models derived using CCBA. Thus, the order of preference for the alignment selection for 3D QSAR model development may be CCBA>DCBA>GMCBA, and the information obtained from the CoMFA and CoMSIA contour maps may be useful in designing specific PTP1B inhibitors.     

SVM-based detection of distant protein structural relationships using pairwise probabilistic suffix trees

A new method based on probabilistic suffix trees (PSTs) is defined for pairwise comparison of distantly related protein sequences. The new definition is adopted in a discriminative framework for protein classification using pairwise sequence similarity scores in feature encoding. The framework uses support vector machines (SVMs) to separate structurally similar and dissimilar examples. The new discriminative system, which we call as SVM-PST, has been tested for SCOP family classification task, and compared with existing discriminative methods SVM-BLAST and SVM-Pairwise, which use BLAST similarity scores and dynamic-programming-based alignment scores, respectively. Results have shown that SVM-PST is more accurate than SVM-BLAST and competitive with SVM-Pairwise. In terms of computational efficiency, PST-based comparison is much better than dynamic-programming-based alignment. We also compared our results with the original family-based PST approach from which we were inspired. The present method provides a significantly better solution for protein classification in comparison with the family-based PST model.     

Stochastic pairwise alignments and scoring methods for comparative protein structure modeling

Despite recent advances in fold recognition algorithms that identify template structures with distant homology to the target sequence, the quality of the target-template alignment can be a major problem for distantly related proteins in comparative modeling. Here we report for the first time on the use of ensembles of pairwise alignments obtained by stochastic backtracking as a means to improve three-dimensional comparative protein models. In every one of the 35 cases, the ensemble produced by the program probA resulted in alignments that were closer to the structural alignment than those obtained from the optimal alignment. In addition, we examined the lowest energy structure among these ensembles from four different structural assessment methods and compared these with the optimal and structural alignment model. The structural assessment methods consisted of the DFIRE, DOPE, and ProsaII statistical potential energies and the potential energy from the CHARMM protein force field coupled to a Generalized Born implicit solvent model. The results demonstrate that the generation of alignment ensembles through stochastic backtracking using probA combined with one of the statistical potentials for assessing three-dimensional structures can be used to improve comparative models.     

An efficient and accurate molecular alignment and docking technique using ab initio quality scoring

An accurate and efficient molecular alignment technique is presented based on first principle electronic structure calculations. This new scheme maximizes quantum similarity matrices in the relative orientation of the molecules and uses Fourier transform techniques for two purposes. First, building up the numerical representation of true ab initio electronic densities and their Coulomb potentials is accelerated by the previously described Fourier transform Coulomb method. Second, the Fourier convolution technique is applied for accelerating optimizations in the translational coordinates. In order to avoid any interpolation error, the necessary analytical formulas are derived for the transformation of the ab initio wavefunctions in rotational coordinates. The results of our first implementation for a small test set are analyzed in detail and compared with published results of the literature. A new way of refinement of existing shape based alignments is also proposed by using Fourier convolutions of ab initio or other approximate electron densities. This new alignment technique is generally applicable for overlap, Coulomb, kinetic energy, etc., quantum similarity measures and can be extended to a genuine docking solution with ab initio scoring.     

CoMFA validation of the superposition of six classes of compounds which block GABA receptors non-competitively

Summary Thirty-six compounds, representing six different structural classes of insecticides which are known to act at the -aminobutyric acid receptor/chloride ionophore, have been superimposed by methods which maximise the commonality of steric and electrostatic fields. Maximal steric and electrostatic alignment was derived by pairwise comparisons of the different chemical classes with picrotoxinin. To test the validity of the combined superposition, a Comparative Molecular Field Analysis (CoMFA) was carried out within SYBYL, using recently published in vivo and in vitro binding data for insecticides. The resultant partial least-squares (PLS) analysis of sampled steric and electrostatic fields showed a significant statistical correlation with the published biological data. The predictive model obtained was shown to have a greater than 95% chance of significance.     

Molecular docking and 3D-quantitative structure activity relationship analyses of peptidyl vinyl sulfones: <Emphasis Type="Italic">Plasmodium Falciparum</Emphasis> cysteine proteases inhibitors

Comparative molecular field analysis (CoMFA) and comparative molecular similarity indices analysis (CoMSIA) based on three-dimensional quantitative structure–activity relationship (3D-QSAR) studies were conducted on a series (39 molecules) of peptidyl vinyl sulfone derivatives as potential Plasmodium Falciparum cysteine proteases inhibitors. Two different methods of alignment were employed: (i) a receptor-docked alignment derived from the structure-based docking algorithm GOLD and (ii) a ligand-based alignment using the structure of one of the ligands derived from a crystal structure from the PDB databank. The best predictions were obtained for the receptor-docked alignment with a CoMFA standard model (q ² = 0.696 and r ² = 0.980) and with CoMSIA combined electrostatic, and hydrophobic fields (q ² = 0.711 and r ² = 0.992). Both models were validated by a test set of nine compounds and gave satisfactory predictive r ² _pred values of 0.76 and 0.74, respectively. CoMFA and CoMSIA contour maps were used to identify critical regions where any change in the steric, electrostatic, and hydrophobic fields may affect the inhibitory activity, and to highlight the key structural features required for biological activity. Moreover, the results obtained from 3D-QSAR analyses were superimposed on the Plasmodium Falciparum cysteine proteases active site and the main interactions were studied. The present work provides extremely useful guidelines for future structural modifications of this class of compounds towards the development of superior antimalarials.     

Geometric accuracy of three-dimensional molecular overlays

This study examines the dependence of molecular alignment accuracy on a variety of factors including the choice of molecular template, alignment method, conformational flexibility, and type of protein target. We used eight test systems for which X-ray data on 145 ligand-protein complexes were available. The use of X-ray structures allowed an unambiguous assignment of bioactive overlays for each compound set. The alignment accuracy depended on multiple factors and ranged from 6% for flexible overlays to 73% for X-ray rigid overlays, when the conformation of the template ligand came from X-ray structures. The dependence of the overlay accuracy on the choice of templates and molecules to be aligned was found to be the most significant factor in six and seven of the eight ligand-protein complex data sets, respectively. While finding little preference for the overlay method, we observed that the introduction of molecule flexibility resulted in a decrease of overlay accuracy in 50% of the cases. We derived rules to maximize the accuracy of alignment, leading to a more than 2-fold improvement in accuracy (from 19% to 48%). The rules also allowed the identification of compounds with a low (<5%) chance to be correctly aligned. Last, the accuracy of the alignment derived without any utilization of X-ray conformers varied from <1% for the human immunodeficiency virus data set to 53% for the trypsin data set. We found that the accuracy was directly proportional to the product of the overlay accuracy from the templates in their bioactive conformations and the chance of obtaining the correct bioactive conformation of the templates. This study generates a much needed benchmark for the expectations of molecular alignment accuracy and shows appropriate usages and best practices to maximize hypothesis generation success.