Robert B. Nachbar

Merck molecular force field. IV. conformational energies and geometries for MMFF94

This article describes the parameterization and performance of MMFF94 for conformational energies, rotational barriers, and equilibrium torsion angles. It describes the derivation of the torsion parameters from high‐quality computational data and characterizes MMFF94s ability to reproduce both computational and experimental data, the latter particularly in relation to MM3. The computational data included: (i) ∼ 250 comparisons of conformational energy based on “MP4SDQ/TZP” calculations (triple‐zeta plus polarization calculations at a defined approximation to the highly correlated MP4SDQ level) at MP2/6‐31G* geometries; and (ii) ∼ 1200 MP2/TZP comparisons of “torsion profile” structures at geometries derived from MP2/6‐31G* geometries. The torsion parameters were derived in restrained least‐squares fits that used the complete set of available computational data, thereby ensuring that a fully optimal set of parameters would be obtained. The final parameters reproduce the “MP4SDQ/TZP” and MP2/TZP computational data with root mean square (rms) deviations of 0.31 and 0.50 kcal/mol, respectively. In addition, MMFF94 reproduces a set of 37 experimental gas‐phase and solution conformational energies, enthalpies, and free energies with a rms deviation of 0.38 kcal/mol; for comparison, the “MP4SDQ/TZP” calculations and MM3 each gives a rms deviation of 0.37 kcal/mol. Furthermore, MMFF94 reproduces 28 experimentally determined rotational barriers with a rms deviation of 0.39 kcal/mol. Given the diverse nature of the experimental conformational energies and rotational barriers and the clear indications of experimental error in some cases, the MMFF94 results appear excellent. Nevertheless, MMFF94 encounters somewhat greater difficulty in handling multifunctional compounds that place highly polar functional groups in close proximity, probably because it, like other commonly used force fields, too greatly simplifies the description of electrostatic interactions. Some suggestions for enhancements to MMFF94s functional form are discussed.

Sample-distance partial least squares: PLS optimized for many variables, with application to CoMFA

SummaryThree-dimensional molecular modeling can provide an unlimited number m of structural properties. Comparative Molecular Field Analysis (CoMFA), for example, may calculate thousands of field values for each model structure. When m is large, partial least squares (PLS) is the statistical method of choice for fitting and predicting biological responses. Yet PLS is usually implemented in a property-based fashion which is optimal only for small m. We describe here a sample-based formulation of PLS which can be used to fit any single response (bioactivity). SAMPLS reduces all explanatory data to the pairwise ‘distances’ among n sample (molecules), or equivalently to an n-by-n covariance matrix C. This matrix, unmodified, can be used to fit all PLS components. Furthermore, SAMPLS will validate the model by modern resampling techniques, at a cost independent of m. We have implemented SAMPLS as a Fortran program and have reproduced conventional and cross-validated PLS analyses of data from two published studies. Full (leaveach-out) cross-validation of a typical CoMFA takes 0.2 CPU s. SAMPLS is thus ideally suited to structure-activity analysis based on CoMFA fields or bonded topology. The sample-distance formulation also relates PLS to methods like cluster analysis and nonlinear mapping, and shows how drastically PLS simplifies the information in CoMFA fields.

Comparison of knowledge-based and distance geometry approaches for generation of molecular conformations.

A knowledge-based approach for generating conformations of molecules has been developed. The method described here provides a good sampling of the molecules conformational space by restricting the generated conformations to those consistent with the reference database. The present approach, internally named et for enumerate torsions, differs from previous database-mining approaches by employing a library of much larger substructures while treating open chains, rings, and combinations of chains and rings in the same manner. In addition to knowledge in the form of observed torsion angles, some knowledge from the medicinal chemist is captured in the form of which substructures are identified. The knowledge-based approach is compared to Blaney et al.s distance geometry (DG) algorithm for sampling the conformational space of molecules. The structures of 113 protein-bound molecules, determined by X-ray crystallography, were used to compare the methods. The present knowledge-based approach (i) generates conformations closer to the experimentally determined conformation, (ii) generates them sooner, and (iii) is significantly faster than the DG method.

Extending the trend vector: the trend matrix and sample-based partial least squares.

SummaryTrend vector analysis [Carhart, R.E. et al., J. Chem. Inf. Comput. Sci., 25 (1985) 64], in combination with topological descriptors such as atom pairs, has proved useful in drug discovery for ranking large collections of chemical compounds in order of predicted biological activity. The compounds with the highest predicted activities, upon being tested, often show a several-fold increase in the fraction of active compounds relative to a randomly selected set. A trend vector is simply the one-dimensional array of correlations between the biological activity of interest and a set of properties or ‘descriptors’ of compounds in a training set. This paper examines two methods for generalizing the trend vector to improve the predicted rank order. The trend matrix method finds the correlations between the residuals and the simultaneous occurrence of descriptors, which are stored in a two-dimensional analog of the trend vector. The SAMPLS method derives a linear model by partial least squares (PLS), using the ‘sample-based’ formulation of PLS [Bush, B.L. and Nachbar, R.B., J. Comput.-Aided Mol. Design, 7 (1993) 587] for efficiency in treating the large number of descriptors. PLS accumulates a predictive model as a sum of linear components. Expressed as a vector of prediction coefficients on properties, the first PLS component is proportional to the trend vector. Subsequent components adjust the model toward full least squares. For both methods the residuals decrease, while the risk of overfitting the training set increases. We therefore also describe statistical checks to prevent overfitting. These methods are applied to two data sets, a small homologous series of disubstituted piperidines, tested on the dopamine receptor, and a large set of diverse chemical structures, some of which are active at the muscarinic receptor. Each data set is split into a training set and a test set, and the activities in the test set are predicted from a fit on the training set. Both the trend matrix and the SAMPLS approach improve the predictions over the simple trend vector. The SAMPLS approach is superior to the trend matrix in that it requires much less storage and CPU time. It also provides a useful set of axes for visualizing properties of the compounds. We describe a randomization method to determine the optimum number of PLS components that is very much faster for large training sets than leave-one-out cross-validation.

Molecular Evolution: Automated Manipulation of Hierarchical Chemical Topology and Its Application to Average Molecular Structures

A simple hierarchical data structure (tree) and associated set of algorithms (written in Mathematica) have been developed that permit the direct manipulation of the topology of a molecule while simultaneously maintaining valid chemical valence. Coupled with a genetic algorithm optimization engine, these computational tools can be used to optimize chemical structures under the guidance of an appropriate fitness function. A detailed study of the factors that influence the performance of the method revealed that it is strongly dependent on the size and complexity of the evolved chemical structures. The effects of population size and choice of genetic operators are much smaller. The results of an exploration into the discovery of average molecular structures using this methodology is also described.

Derivation of a 3D pharmacophore model for the angiotensin-II site one receptor.

SummaryA systematic search has been used to derive a hypothesis for the receptor-bound conformation of A-II antagonists at the AT1 receptor. The validity of the pharmacophore hypothesis has been tested using CoMFA, which included 50 diverse A-II antagonists, spanning four orders of magnitude in activity. The resulting cross-validated R2 of 0.64 (conventional R2 of 0.76) is indicative of a good predictive model of activity, and has been used to estimate potency for a variety of non-peptidyl antagonists. The structural model for the non-peptide has been compared with respect to the natural substrate, A-II, by generating peptide to non-peptide overlays.

The design, binding affinity prediction and synthesis of macrocyclic angiotensin II AT1 and AT2 receptor antagonists

Analysis of the SAR of AT1 selective and AT1AT2 balanced affinity angiotensin II antagonists led to the design of macrocyclic quinazolinone ligands. CoMFA analysis was used to predict the binding affinities of these novel ligands. The synthesis, X-ray crystal structure, binding affinity and the relevance of these studies to the determination of the biologically relevant binding conformation is discussed.

Quinazolinones 2: QSAR and in vivo characterization of AT1 selective AII antagonists

Abstract The structure activity relatoionship, linear regression analysis and in vivo evaluation of a series of substituted 2-butyl-3-[(2′-tetrazol-5-yl)biphen-4-yl)methyl]quinazolin-4(1H)-ones as antagonists of the AT1 receptor for angiotensin II is presented. L-159,093 (2-butyl-6-(N-isopropyl-N-methyl-carbamoyl)amino-3-[(2′-tetrazol-5-yl)biphen-4-yl)methyl]quin azolin-4(1H)-one (IC50=0.1nM rabbit aorta) is shown to be a potent orally active AII antagonist in rats and rhesus.

Modeling Effect of a γ-Secretase Inhibitor on Amyloid-β Dynamics Reveals Significant Role of an Amyloid Clearance Mechanism

Aggregation of the small peptide amyloid beta (Aβ) into oligomers and fibrils in the brain is believed to be a precursor to Alzheimer’s disease. Aβ is produced via multiple proteolytic cleavages of amyloid precursor protein (APP), mediated by the enzymes β- and γ-secretase. In this study, we examine the temporal dynamics of soluble (unaggregated) Aβ in the plasma and cerebral-spinal fluid (CSF) of rhesus monkeys treated with different oral doses of a γ-secretase inhibitor. A dose-dependent reduction of Aβ concentration was observed within hours of drug ingestion, for all doses tested. Aβ concentration in the CSF returned to its predrug level over the monitoring period. In contrast, Aβ concentration in the plasma exhibited an unexpected overshoot to as high as 200% of the predrug concentration, and this overshoot persisted as late as 72 hours post-drug ingestion. To account for these observations, we proposed and analyzed a minimal physiological model for Aβ dynamics that could fit the data. Our analysis suggests that the overshoot arises from the attenuation of an Aβ clearance mechanism, possibly due to the inhibitor. Our model predicts that the efficacy of Aβ clearance recovers to its basal (pretreatment) value with a characteristic time of >48 hours, matching the time-scale of the overshoot. These results point to the need for a more detailed investigation of soluble Aβ clearance mechanisms and their interaction with Aβ-reducing drugs.

COMPARISON OF CONSTITUTIONAL PROPERTIES IN THE CIP SYSTEM: THE CIP SEQUENCE RULE 1

Abstract An analysis of the CIP Sequence Rules was recently published in this journal 1 . However, further work on the implementation of the CIP Rules for computer use suggested that a reformulation of Sequence Rule 1 was required. In this paper this Rule is analyzed and its modification proposed.