Search PubMed⌕ Search

Biomedical subjects

D A Winkler

Publications and source records attributed to D A Winkler.

13 recordsLinked to original sources

A quantitative structure--activity relationships model for the acute toxicity of substituted benzenes to Tetrahymena pyriformis using Bayesian-regularized neural networks.

We have used a new, robust structure-activity mapping technique, a Bayesian-regularized neural network, to develop a quantitative structure-activity relationships (QSAR) model for the toxicity of 278 substituted benzenes toward Tetrahymena pyriformis. The independent variables used in the modeling were derived solely from the molecular structure, and the model was tested on 20% of the data set selected from the whole set by cluster analysis and which had not been used in training the network. The results show that the method is robust and reliable and give results for mixed class compounds which are comparable to earlier QSAR work on single-chemical class subsets of the 278 compounds and which employed measured physicochemical parameters as independent variables. Comparisons of Bayesian neural net models with those derived by classical PLS analysis showed the superiority of our method. The method appears to be able to model more diverse chemical classes and more than one mechanism of toxicity.

Animals↗

Robust QSAR models using Bayesian regularized neural networks.

We describe the use of Bayesian regularized artificial neural networks (BRANNs) in the development of QSAR models. These networks have the potential to solve a number of problems which arise in QSAR modeling such as: choice of model; robustness of model; choice of validation set; size of validation effort; and optimization of network architecture. The application of the methods to QSAR of compounds active at the benzodiazepine and muscarinic receptors is illustrated.

Drug Design↗

Molecular modeling studies of "flap up" mannosyl cation mimics.

The importance of inhibitors of glycosidases as therapeutic agents for viral, proliferative, and metabolic diseases is being increasingly recognised. Several years ago we reported that the activities of mannosidase inhibitors may be explained in terms of their similarity to the mannosyl cation intermediate postulated to form during the enzyme-catalyzed processing of oligosaccharide substrates. Recently, the validity of this model has been called in to question by some authors. We report recent molecular modeling studies undertaken to clarify this apparent contradiction. Mannostatin can indeed bind in a fashion which bears a close similarity to the mannosyl cation. Moreover we have shown that (-)-mannostatin is not able to adopt a similar binding mode to that of the mannosyl cation. As additional proof, Farr et al. have synthesized a trihydroxycyclopentylamine as a direct mimic of our mannosyl cation model. Satisfyingly, this compound shows potent inhibition of Jack Bean alpha-mannosidase, as predicted by the model. The inactivity of aminotrihydroxyhexahydro-1H-azepine against mannosidases can be explained in terms of the relative energies of the axial versus equatorial conformations of the critical hexahydroazepine ring substituents.

Azepines↗

Morpheus: a conformation-activity relationships and receptor modeling package.

Our molecular modeling software package, MORPHEUS, allows the study of the interactions between biologically active molecules and their receptors. The package is capable of exploring the multidimensional conformational space accessible to each molecule of the data set under study. By specifying distance constraints or hypothetical receptor binding points, the package is able to filter the biologically accessible conformations of each active compound and deduce a three-dimensional model of the binding sites consistent with the properties of the agonists (or antagonists) under scrutiny. The electrostatic potentials in the environment of a putative binding site can also be investigated using the MORPHEUS package. The molecular modeling module CRYS-X, which is written in FORTRAN 77 for IBM PC machines, is capable of building, displaying and manipulating molecules.

Computer Graphics↗

Design of potential anti-HIV agents. 1. Mannosidase inhibitors.

A molecular orbital and molecular graphics study of 12 substrates, inhibitors, reaction intermediates, and substrate analogues of alpha-mannosidase was undertaken. The results indicated that potent inhibitors must be good topographical analogues of the mannopyranosyl cation, an intermediate in the reaction catalyzed by the enzyme. Enzyme recognition and strong binding by the inhibitors requires that they contain, as part of their structure, electronegative atoms which are the topographical equivalent of the mannosyl cation C2 and C3 hydroxyl groups and ring heteroatom. The absence of a topographical analogue of the C4 hydroxyl group of the cation appeared to have little effect on the binding and activity of inhibitors. These results have been utilized in the design of potential anti-HIV drugs whose synthesis is now under consideration.

Alkaloids↗

Modern mammal origins: evolutionary grades in the Early Cretaceous of North America.

Major groups of modern mammals have their origins in the Mesozoic Era, yet the mammalian fossil record is generally poor for that time interval. Fundamental morphological changes that led to modern mammals are often represented by small samples of isolated teeth. Fortunately, functional wear facets on teeth allow prediction of the morphology of occluding teeth that may be unrepresented by fossils. A major step in mammalian evolution occurred in the Early Cretaceous with the evolution of tribosphenic molars, which characterize marsupials and placentals, the two most abundant and diverse extant groups of mammals. A tooth from the Early Cretaceous (110 million years before present) of Texas tests previous predictions (based on lower molars) of the morphology of upper molars in early tribosphenic dentitions. The lingual cusp (protocone) is primitively without shear facets, as expected, but the cheek side of the tooth is derived (advanced) in having distinctive cusps along the margin. The tooth, although distressingly inadequate to define many features of the organism, demonstrates unexpected morphological diversity at a strategic stage of mammalian evolution and falsifies previous claims of the earliest occurrence of true marsupials.

Animals↗

NMR and molecular orbital studies of isomerism and tautomerism in oximes of 2-acyl cyclic 1,3-diones.

Molecular orbital calculations and 13C and 15N NMR experiments have been performed on the O-methyl oximes of two types of 2-acyl cyclic 1,3-diones. One (III; X = CH2) was based on cyclohexane, and the other (X = O) on pyran. The data indicate that both the cyclohexane and pyran compounds prefer to exist as the oxime, rather than the enamine isomer. Two equivalent, interconverting, keto-enol tautomers exist in the cyclohexane compound. The pyran compound exists as the lactone-enol tautomer. Our results have implications in the design of herbicidal compounds and drugs containing similar tautomeric systems.

Cyclohexanes↗

QSAR studies of hydrazone uncouplers of oxidative phosphorylation.

Semiempirical molecular orbital calculations have been performed on a series of hydrazone uncouplers of mitochondrial oxidative phosphorylation which show insecticidal activity. Regression analysis yielded significant correlations between uncoupling activity, insecticidal potency and such physicochemical or theoretically-derived parameters as lipophilicity, pKa and atom charges.

Adenosine Diphosphate↗

Conformational energy calculations and electrostatic potentials of dihydrofolate reductase ligands: relevance to mode of binding and species specificity.

Classical potential energy calculations are reported for a series of 11 structurally diverse substrates, products, and inhibitors of dihydrofolate reductase. In almost every case, the calculations reveal a range of potential biologically active conformations accessible to the molecule, and geometry optimization with molecular mechanics and molecular orbital calculations further expands the range of accessible conformations. The energy calculations are supplemented with electrostatic potential energy surfaces for the heterocyclic components of each molecule. These data are used in conjunction with the energy calculations and the crystallographically determined enzyme structures to compare two alternative proposed binding modes of folates known to bind with their pteridine rings inverted relative to that of methotrexate. It is shown that the conformational flexibility of the connecting chain between the benzoyl glutamate and pteridine moieties in the folates actually allows the pteridine ring to shift between these alternative binding modes, a combination of which may offer the best explanation for the observed activity. The electrostatic potentials and conformational energy data are also used in an attempt to account for the species specificity of inhibitors of mammalian, bacterial, and protozoal dihydrofolate reductases. The results show that while these techniques can be used to explain many of the observed results, others require recourse to the observed crystal structures to provide a satisfactory explanation.

Aminopterin↗

Structure-activity relationships of convulsant and anticonvulsant barbiturates: a computer-graphic-based pattern-recognition analysis.

A computer-graphic-based pattern-recognition study of two series of 5-ethyl-5-substituted barbiturates has been undertaken in an attempt to find a correlation between molecular conformation and convulsant and anticonvulsant activity. Studies of a first (trial) set of barbiturates related to pentobarbital revealed a region of space in which at least one low-energy conformation of the hydrocarbon side chain of each of the anticonvulsant barbiturates resides. Another region was occupied by a low-energy conformation of each of the convulsant barbiturates. These regions of space are, thus, possible pharmacophores for convulsant and anticonvulsant activity. Analysis of a second (test) set of barbiturates related to phenobarbital has shown that the activities and structures of these molecules are consistent with the above model. These pharmacophores thus provide a basis for the design of rigid, new analogues with potent convulsant or anticonvulsant activities.

Anticonvulsants↗

Conformational analysis of the ergot alkaloids ergotamine and ergotaminine.

Conformational analyses by 1H NMR and potential-energy calculations are reported for the ergot alkaloids ergotamine and ergotaminine, both as free bases and as the protonated species. In the neutral forms in CDCl3. two strong intramolecular hydrogen bonds fix the molecules in folded conformations, but the protonated species adopt a more extended conformation, with a single intramolecular hydrogen bond. Of the 24 alternative conformations available to ergotamine, the most likely biologically active species in environments with low dielectric constants, e.g., the presumed ergotamine binding site, is the folded, hydrogen-bonded conformation observed for the neutral molecule in CDCl3 solution.

Chemical Phenomena↗

Use of automatic relevance determination in QSAR studies using Bayesian neural networks.

We describe the use of Bayesian regularized artificial neural networks (BRANNs) coupled with automatic relevance determination (ARD) in the development of quantitative structure-activity relationship (QSAR) models. These BRANN-ARD networks have the potential to solve a number of problems which arise in QSAR modeling such as the following: choice of model; robustness of model; choice of validation set; size of validation effort; and optimization of network architecture. The ARD method ensures that irrelevant or highly correlated indices used in the modeling are neglected as well as showing which are the most important variables in modeling the activity data. The application of the methods to QSAR of compounds active at the benzodiazepine and muscarinic receptors as well as some toxicological data of the effect of substituted benzenes on Tetetrahymena pyriformis is illustrated.

Journal Article↗