摘要:

AbstractA QSAR analysis of the [3H]spiperone binding affinity of 25 mono‐ and disubstituted salicylamides having F, Cl, Br, Me and Et substituents in the 3‐ and 5‐positions have been done with PLS (partial least squares in latent variables) method. Besides descriptors for size and lipophilicity, a large number of theoretically derived descriptors, i.e. polarizability, charge and molecular electrostatic potential (MEP), were used. The partial charges and the 3‐D MEPs were calculated using GAUSSIAN 80 at the STO‐3G level on model compounds lacking the pyrrolidine side‐chain. The MEPs at the van der Waals distance and at a larger distance corresponding to 3.0 Å from the aromatic plane were converted to point charges centered at the heavy atoms. The main component t1explains 71% of the observed variance in biological activity and with inclusion of two minor components t2and t3a total of 88% explained variance is obtained. The size, lipophilicity and electronics of the 3‐substituent are of major importance. The MEP descriptors at the 3.00 Å distance also influence the activity despite the fact that the gross behaviour of the 2‐D MEPs was highly variable with no apparent correlation with D‐2 activity. The present study does not fully support current MEP pharmacophores on closel

ISSN:0931-8771    

DOI:10.1002/qsar.19910100102

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractThe concept and methodology of artificial neural networks is introduced. Like pattern recognition, the techniques can be classified as supervised (requiringa prioriknowledge of class membership) and unsupervised (making no assumptions about class membership). An unsupervised neural network method, Kohonen Topology‐Preserving Mapping, is applied to a wide matrix of physicochemical property data for a set of antifilarial antimycin analogues containing structural outliers. Principal component analysis failed to give a good 2D representation of the data set as a whole due to linear constraints in the model which gave undue influence to the outliers. Kohonen mapping compared favourably with non‐linear unsupervised statistical pattern recognition methods for 2D representation of compound similarity and for classification based on antifilarial activity. It may prove a valuable technique for QSAR in situations where a linear method does not model the data well and a high throughput of test compounds is indica

ISSN:0931-8771    

DOI:10.1002/qsar.19910100103

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractA separate consideration of ligand‐β‐adrenoceptor interactions responsible for affinity (recognition, binding) and agonistic activity, respectively, is very important for the explanation of molecular events eliciting the activating conformational change of the receptor. Equilibrium models of β‐adrenergic effectuation including receptor activation by agonists and partial agonists, receptor‐effector (G‐protein) coupling, and adenylate cyclase stimulation allow to obtain LFER‐related parameters of both interaction components from apparent dissociation constants and intrinsic activities. Two models, both containing “precoupling” of non‐activated receptor and effector molecules, are derived: a ternary complex model for the interaction of ligands, receptor, and effector in the absence, and an adenylate cyclase activation model for enzyme activation in the presence of GTP. Agonistic activity is represented by the equilibrium constant K* reflecting the concentration ratio of activated and non‐activated agonist‐receptor complexes. KD, the dissociation constant of non‐activated complexes, represents ligand‐receptor recognition and binding. K* can be estimated either from using ratios KL/KHof apparent low‐ and high affinity dissociation constants in GTP absence, or from using intrinsic activities of adenylate cyclase activation. According to our models, KL, which also describes apparent dissociation in binding assays with GTP, is related to KDand K* with KL= KD/(1 + K*). In terms of the free energy of ligand binding and receptor activation. ΔG° = RT ln KL, this corresponds to a decomposition of pKLinto pKL= pKD+ log (1 + K*). In QSAR, thus, log (1 + K*) is the parameter of choice for LFER‐related description of ligand‐receptor intera

ISSN:0931-8771    

DOI:10.1002/qsar.19910100104

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractBased on an equilibrium model of ligand‐β‐adrenoceptor interactions and β‐adrenergic effectuation presented in a preceding paper, LFER‐related biological parameters to characterize affinity to the receptor and agonistic potency, respectively, are derived for a series of 44 phenethanolamines from measured apparent dissociation constants, KL, and intrinsic activities αACof adenylate cyclase activation. Both parameters result from a decomposition of pKLinto pKL= pKD+ log (1 + K*). Fragment regression analysis of pKLwas used for a preceding separation of the maximal contribution of agonistic interactions to apparent affinity. The log (1 + K*) scale was then estimated as logit transformation of αAC. Hansch analyses of log (1 + K*) and pKDlead to a clear separation of molecular features like substituent positions and interaction types responsible for agonistic potency and affinity, respectively. Thus, both parameters are proved to be of practical value, particularly as the resulting regression coefficients in Hansch equations are directly related to energy contributions of “basic” int

ISSN:0931-8771    

DOI:10.1002/qsar.19910100105

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractThe 9,9‐bis(4‐fluorophenyl)‐3,5‐dihydroxy‐8‐(substituted)‐6, 8‐nonadienoic acid analogues represent a novel class of HMG‐CoA reductase inhibitors developed in our laboratories. We delineate from inhibitory potency values the main topographical and physicochemical features of the binding site probed by substituents attached to the C8position of the analogues. Using a combination of receptor mapping and 3D‐QSAR techniques, it was possible to determine a logical candidate for the conformation of the inhibitor bound to the receptor and to derive a reliable 3D‐QSAR which relates the HMG‐CoA reductase inhibitory potency to the shape and size of both the binding site and C8‐substituent of the inhibitor. We show that the 3D‐QSAR derived here

ISSN:0931-8771    

DOI:10.1002/qsar.19910100106

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractEight chemical model systems were developed for a multivariate physicochemical characterization of halogenated aliphatics. The chemical model systems were developed to obtain descriptors which contained information related to lipophilicity and chemical reactivity of the studied compounds. A selected set of 15 halogenated aliphatic compounds were characterized by the chemical model systems. The information content in eight new descriptors combined with eight old physico‐chemical descriptors were validated in a QSAR. The used biological responses were acute oral toxicity to rat, highest non‐lethal dose to mouse and in vitro genotoxicity from a DNA precipitation assay. This model was compared with a QSAR based on eight old‐physico‐chemical descriptors. It was found that the addition of the new descriptors improved t

ISSN:0931-8771    

DOI:10.1002/qsar.19910100107

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

摘要:

AbstractA new characterization of atoms in molecules is introduced as the electrotopological state index, which combines both the electronic character and the topological environment of each skeletal atom in a molecule. The electrotopological state (E‐state) of a skeletal atom is formulated as an intrinsic value Iiplus a perturbation term ΔIi, arising from the electronic interaction within the molecular topological environment of each atom in the molecule. The atom intrinsic value, for first row atoms, is expressed as I = (δv+ 1)δ, in which δvand δ are the counts of valence and sigma electrons, respectively, for the atom in the molecular skeleton. The E‐state, Si, for atom i is defined as Si= Ij+ ΔIj, where the influence of other atoms on atom i, ΔIi, is given as Σ(Ij— Ij)/rij2; rijis the graph separation between atoms i and j, counted as number of atoms, including i and j. Information in the electrotopological state is revealed by examples of various types of organic structures, including skeletal branching and heteroatom variation. The relation of the E‐state value to NMR chemical shift is demonstrated for a series of carbonyl compounds. QSAR examples are given for hydrazide inhibition of MAO and for receptor binding of β‐carbolines. These examples reveal the power of this approach to QSAR using atom level indexes, computed directly from molecule connection tables, in which it is possible to identify atoms and regions in the molecule which are impo

ISSN:0931-8771    

DOI:10.1002/qsar.19910100108

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

ISSN:0931-8771    

DOI:10.1002/qsar.19910100109

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

ISSN:0931-8771    

DOI:10.1002/qsar.19910100110

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY

ISSN:0931-8771    

DOI:10.1002/qsar.19910100111

出版商:WILEY‐VCH Verlag    

年代:1991

数据来源: WILEY