Molecular quantum similarity in QSAR and drug design
著者
書誌事項
Molecular quantum similarity in QSAR and drug design
（Lecture notes in chemistry, 73）
Springer-Verlag, c2000
大学図書館所蔵 件 / 全7件
-
該当する所蔵館はありません
- すべての絞り込み条件を解除する
注記
Includes bibliographical references
内容説明・目次
内容説明
The studyofthehealingpowerofchemicalcompounds, present into the known natural active principles and its subsequent use, shall be seen initially as a pseudoscientificprocedure,acontinuationoftheChineseandArabianoccultism, whichwasbased inthepercentagecontentoffire,air,earthandwater,aswellas ontheassociatedqualities:hot-cold,humid-dry...,whichthematterwassupposed to be formed by. Themethod, possessingroots in Hippocrates, Dioscoridesand Galen was studied, described and polished by Avicenna, Averrot!s, Bacon and Villanova inthe MiddleAge. Italsowasappearingasastudyconstantduringthe renaissance and after. On the other hand, the birthofchemistry asascientific offspring from alchemy propitiated alternative ways ofknowledge in order to solve the same problem. In this manner, in the past century, approximately hundred years from now, Sylvester proposed the first molecular description in numerical discrete form, employing ideas which even in present times can be associated withinthesocalledmoleculartopology. Sylvester'stopologicalmodel can be considered the seed allowing the originofthis big tree, which is now knownastheoreticalchemistry. .
During all the past time from the first topological modelofSylvester up to now, the proliferationofnumerical parameters to describe molecular structures has not ceased to grow larger. Some ofthese parameters have played a very important role for the understandingofthe organic molecules behavior and, by extension, for the comprehension and evaluation of their physical as well as biologicalproperties. InthemindofeveryspecialistaretheHammett'scr,theTaft constantsor the octanol-water partition coefficient. Other numerical parameters, suchasthosederivedfromthemodemtopologicalmolecularrepresentationarein aprocessofconstantrevisionandgrowing. Thus,theHosoyaandRandicindices, ortheKier'sconnectivities,amongseveralnotsowellknownnumericaldataare usual reference descriptors. They are putatthe researchers' disposition,andare easily deducible from any molecular representation in form ofordered setsof numerical figures. All ofthem are profusely studied and employed in present times.
The main idea consists into the useofthese numerical data in orderto obtaininformationonthemoleculartrendstopossessoracquirecertainproperties and, even better than this, to determine in which degree or intensity molecules presenteverything.
目次
1 Introduction.- 1.1 Origins and evolution of QSAR.- 1.2 Molecular similarity in QSAR.- 1.3 Scope and contents of the book.- 2 Quantum objects, density functions and quantum similarity measures.- 2.1 Tagged sets and molecular description.- 2.1.1 Boolean tagged sets.- 2.1.2 Functional tagged sets.- 2.1.3 Vector semispaces.- 2.2 Density functions.- 2.3 Quantum objects.- 2.4 Expectation values in Quantum Mechanics.- 2.5 Molecular Quantum Similarity.- 2.6 General definition of molecular quantum similarity measures (MQSM).- 2.6.1 Overlap MQSM.- 2.6.2 Coulomb MQSM.- 2.7 Quantum self-similarity measures.- 2.8 MQSM as discrete matrix representations of the quantum objects..- 2.9 Molecular quantum similarity indices (MQSI).- 2.9.1 The Carbo index.- 2.10 The Atomic Shell Approximation (ASA).- 2.10.1 Promolecular ASA.- 2.10.2 ASA parameters optimization procedure.- 2.10.3 Example of ASA fitting: adjustment to ab initio atomic densities using a 6-31 IG basis set.- 2.10.4 Descriptive capacity of ASA.- 2.11 The molecular alignment problem.- 2.11.1 Dependence of MQSM with the relative orientation between two molecules.- 2.11.2 Maximal similarity superposition algorithm.- 2.11.3 Common skeleton recognition: the topo-geometrical superposition algorithm.- 2.11.4 Other molecular alignment methods.- 3 Application of Quantum Similarity to QSAR.- 3.1 Theoretical connection between QS and QSAR.- 3.1.1 Beyond the expectation value.- 3.2 Construction of the predictive model.- 3.2.1 Multilinear regression.- 3.3 Possible alternatives to the multilinear regression.- 3.3.1 Partial least squares (PLS) regression.- 3.3.2 Neural Network algorithms.- 3.4 Parameters to assess the goodness-of-fit.- 3.4.1 The multiple determination coefficient r2.- 3.4.2 The standard deviation coefficient ?N.- 3.5 Robustness of the model.- 3.5.1 Cross-validation by leave-one-out.- 3.5.2 The prediction coefficient q2.- 3.5.3 Influence on the regression results.- 3.6 Study of chance correlations.- 3.6.1 The randomization test.- 3.7 Comparison between the QSAR models based on MQSM and other 2D and 3D QSAR methods.- 3.7.1 Comparison with 2D methods.- 3.7.2 Comparison with 3D methods built on grids.- 3.8 Limitations of the models based on MQSM.- 3.8.1 Homogeneity of the sets.- 3.8.2 The problem of the bioactive conformation.- 3.8.3 Determination of molecular alignment.- 4 Full molecular quantum similarity matrices as QSAR descriptors.- 4.1 Pretreatment for quantum similarity matrices.- 4.1.1 Dimensionality reduction.- 4.1.2 Variable selection.- 4.2 The MQSM-QSAR protocol.- 4.3 Combination of quantum similarity matrices: the tuned QSAR model.- 4.3.1 Mixture of matrices and coefficient constraints.- 4.3.2 Optimization of the convex coefficients.- 4.4 Examples of QSAR analyses from quantum similarity matrices.- 4.4.1 Activity of indole derivatives.- 4.4.2 Aquatic toxicity of substituted benzenes.- 4.4.3 Single-point mutations in the subtilisin enzyme.- 5 Quantum self-similarity measures as QSAR descriptors.- 5.1 Simple QSPR models based on QS-SM.- 5.2 Characterization of classical 2D QSAR descriptors using QS-SM.- 5.2.1 QS-SM as an alternative to log P values.- 5.2.2 QS-SM as an alternative to Hammett a constant.- 5.3 Description of biological activities using fragment QS-SM.- 5.3.1 Activity against Bacillus cereus ATCC 11778 (Bc).- 5.3.2 Activity against Streptococcus faecalis ATCC 10541 (Sf).- 5.3.3 Activity against Staphylococcus aureus ATCC 25178 (Sa).- 6 Electron-electron repulsion energy as a QSAR descriptor.- 6.1 Connection between the electron-electron repulsion energy and QS-SM.- 6.2 ?Vee? as a descriptor for simple linear QSAR models.- 6.3 Evaluation of molecular properties using ?Vee? as a descriptor.- 6.3.1 Inhibition of spore germination by aliphatic alcohols.- 6.3.2 Inhibition of microbial growth by aliphatic alcohols and amines.- 6.3.3 Aquatic toxicity of benzene-type compounds.- 6.3.4 Activity of alkylimidazoles.- 7 Quantum similarity extensions to non-molecular systems: Nuclear Quantum Similarity.- 7.1 Generality of Quantum Similarity for quantum systems.- 7.2 Nuclear Quantum Similarity.- 7.2.1 Nuclear density functions: the Skyrme-Hartree-Fock model.- 7.3 Structure-property relationships in nuclei.- 7.3.1 The nuclear data set.- 7.3.2 The binding energy per nuclcon.- 7.3.3 The mass excess.- 7.4 Limitations of the approach.- References.
「Nielsen BookData」 より