Clever Geek Handbook
📜 ⬆️ ⬇️

QSAR

The search for quantitative structure-property relationships is a procedure for constructing models that allow predicting their various properties by the structures of chemical compounds. For models to predict the quantitative characteristics of biological activity, the English name Quantitative Structure-Activity Relationship (QSAR) has historically entrenched. The acronym QSAR is often interpreted extensively to refer to any structure-property models. The models, allowing to predict the physical and physicochemical properties of organic compounds, have the English name Quantitative Structure-Property Relationship (QSPR) . For a qualitative description of the relations between the structures of chemical compounds and their biological activity, the English term Structure-Activity Relationship (SAR) is used .

The search for quantitative relationships between structure and property is based on the application of mathematical statistics and machine learning methods to build models that allow predicting their properties (physical, chemical, biological activity) by describing the structures of chemical compounds. When predicting properties at a qualitative level (for example, whether a given chemical compound will possess this type of biological activity), they talk about solving the classification problem, while when predicting the numerical values ​​of properties, they talk about solving a regression problem. The description of the structures of chemical compounds for these purposes can be vector or non-vector ( graph ).

Modeling properties in the vector description of chemical compounds

In the vector description of the chemical structure, the vector of molecular descriptors is assigned, each of which is an invariant of the molecular graph .

Molecular Descriptors

Existing sets of molecular descriptors can be conditionally divided into the following categories:

  1. Fragment descriptors [1] [2] [3] exist in two main versions - binary and integer . Binary fragment descriptors indicate whether a given fragment (substructure) is contained in a structural formula (i.e., whether a given subgraph is contained in a molecular graph describing a given chemical compound ), while integer fragment descriptors indicate how many times a given fragment (substructure) is contained in a structural formula (that is, how many times a given subgraph is contained in a molecular graph describing a given chemical compound ). The review [3] describes 11 main categories of fragment descriptors. The unique role of fragment descriptors is that, as shown in [4] [5] , they form the basis of the descriptor space, that is, any molecular descriptor (and any molecular property) that is an invariant of the molecular graph can be uniquely decomposed according to this basis. In addition to modeling the properties of organic compounds, binary fragment descriptors in the form of molecular keys (screenshots) and molecular fingerprints are used when working with databases to accelerate substructural search and organize similar searches [3] .
  2. Topological indices .
  3. Physicochemical descriptors [6] are numerical characteristics obtained by modeling the physicochemical properties of chemical compounds, or quantities that have a clear physicochemical interpretation. Most commonly used as descriptors: lipophilicity (LogP), molar refraction (MR), molecular weight (MW), hydrogen bonding descriptors [7] , molecular volumes and surface areas.
  4. Quantum chemical descriptors [8] are numerical values ​​obtained as a result of quantum chemical calculations. The most commonly used descriptors are: energies of boundary molecular orbitals (HOMO and LUMO), partial charges on atoms and partial bond orders, Fukui reactivity indices (free valence index, nucleophilic and electrophilic superdelocalizability), cationic, anionic and radical localization energies, dipole and higher multipole moments of the distribution of electrostatic potential.
  5. Molecular field descriptors are numerical values ​​that approximate the values ​​of molecular fields by calculating the interaction energy of a test atom placed in a lattice site with the current molecule. The construction of correlations between the values ​​of molecular field descriptors and the numerical value of biological activity using the Partial Least Squares (PLS ) method is based on 3D-QSAR methods, the most famous of which is CoMFA [9] .
  6. The substituent constants [10] were first introduced by L. P. Hammett in the framework of the equation that received his name, which relates the reaction rate constants to the equilibrium constants for some classes of organic reactions. The substituent constants entered QSAR practice after the appearance of the Gancha-Fujita equation, which relates the biological activity with the substituent constants and the lipophilicity value. Currently, several dozen constants of substituents are known.
  7. Pharmacophore descriptors show whether the simplest pharmacophores, consisting of pairs or triples of pharmacophore centers with a specified distance between them, can be contained inside the analyzed molecule [11] .
  8. Descriptors of molecular similarity indicate a measure of similarity (molecular similarity) with compounds from the training set.

Molecular descriptors are most fully described in the monograph [12] , which can be considered the encyclopedia of molecular descriptors, as well as in the textbook [13] .

Methods for constructing structure-property models

To solve regression problems in the vector description of the structures of chemical compounds , the following methods of mathematical statistics and machine learning are most often used in chemoinformatics:

  1. Multiple linear regression
  2. Partial Least Squares (PLS )
  3. Artificial Neural Networks
  4. Regression on support vectors
  5. Random forest
  6. Nearest Neighbor Method k

To solve two-class (binary) or multiclass classification problems in the vector description of the structures of chemical compounds , the following methods of mathematical statistics and machine learning are most often used in chemoinformatics:

  1. Naive Bayes Classifier
  2. Linear Discriminant Analysis ( LDA )
  3. Artificial Neural Networks
  4. Support Vector Method
  5. Decision trees
  6. Random forest
  7. Nearest Neighbor Method k

To solve one-class classification problems in the vector description of the structures of chemical compounds , the following machine learning methods are most often used in chemoinformatics:

  1. Autocoding Neural Networks
  2. One Class Support Vector Machine (1-SVM)

Methods for constructing structure-property models are discussed in detail in textbooks [13] [14] .

Modeling of properties with non-vector (graph) description of chemical compounds

Modeling of properties in the non-vector description of chemical compounds is carried out either using neural networks of special architectures that allow working directly with adjacency matrices of molecular graphs , or using nuclear (kernel) methods using special graph (or chemical, pharmacophore) nuclei.

Examples of employees for this purpose neural networks with a special architecture are:

  1. BPZ [15] [16]
  2. ChemNet [17]
  3. CCS [18] [19]
  4. MolNet [20]
  5. Graph machines [21]

Examples of graph (or chemical, pharmacophore) nuclei serving for this purpose are:

  1. Marginalized graph kernel [22]
  2. Optimal assignment kernel [23] [24] [25]
  3. Pharmacophore kernel [26]

The construction of structure-property models for non-vector descriptions of chemical structures is considered in the manual [14] .

Computing resources freely available over the Internet

Resources for building new structure-property models

  1. Online CHemical Modeling (OCHEM) is an informational and computational resource that allows you to work with a database of organic compounds and their properties via the Web interface, replenish it, search and form samples in it, calculate a wide range of molecular descriptors, build quantitative models of structure- property and apply them to predict the properties of new compounds
  2. Chembench - a resource that allows you to build structure-property models and use them for forecasting

Forecasting Examples

Examples of predicting the physicochemical properties of organic compounds

  1. Physical properties of individual low molecular weight compounds
    1. Boiling point (Tk) [27] [28]
    2. Critical Temperature (Tcr) [27]
    3. Viscosity [28] [29]
    4. Saturated vapor pressure [27] [28] [29]
    5. Density [27] [28] [29]
    6. Refractive index [27]
    7. Melting point ( mp ) [27]
    8. Solvent polarity scales [27]
    9. Retention indices in gas chromatography [27]
    10. Polarizability [30]
    11. Magnetic susceptibility [31]
    12. Sublimation Enthalpy [32]
  2. Physical properties of low molecular weight compounds depending on conditions
    1. The boiling point of hydrocarbons as a function of pressure [33]
    2. Density of hydrocarbons as a function of temperature [33]
    3. Dynamic viscosity of hydrocarbons as a function of temperature [33]
  3. Spectroscopic properties
    1. The position of the long-wavelength absorption band of symmetric cyanine dyes [34]
    2. Chemical shifts in the 1 H NMR spectra [35]
    3. Chemical shifts in the 13 C NMR spectra [36]
    4. Chemical shifts in the 31 P NMR spectra [37]
  4. Physical and chemical properties of low molecular weight compounds
    1. Flash point and auto-ignition temperature [27] [38]
    2. Hydrocarbon octane numbers [39]
    3. Ionization constants ( acidity or basicity ) [40]
  5. Physical properties due to intermolecular interactions of molecules of various types
    1. Solubility in Water (LogSw) [27] [41]
    2. Partition coefficient n-octanol / water (LogP) [42]
    3. Partition coefficient of low molecular weight substances between water and micelles Pluronic P85 [43]
    4. Free energy of solvation of organic molecules in various solvents [44]
  6. Reactivity of Organic Compounds
    1. The rate constant of acid hydrolysis of esters [45]
  7. Supramolecular properties
    1. Stability of inclusion complexes of organic compounds with beta-cyclodextrin [46]
    2. The affinity of dyes to cellulose fiber [47]
    3. The stability constants of complexes of ionophores with metal ions [48]
  8. Physical properties of surfactants
    1. Critical Micelle Concentration (CMC) [27]
    2. Cloud Point [27]
  9. Physical and physico-chemical properties of polymers
    1. Glass transition temperature [27]
    2. The refractive index of polymers [27]
    3. Acceleration of rubber vulcanization [27]
    4. Permeability coefficient through low density polyethylene [49]
  10. Physical properties of ionic liquids
    1. Melting point [50]

ADMET Property Prediction Examples

  1. Pharmacokinetic properties
    1. Penetration through the blood-brain barrier [51]
    2. Skin penetration rate [52]
  2. Metabolism
    1. Aromatic hydroxylation sites during metabolic activation by cytochrome P450 [53]
  3. Toxicity
    1. Carcinogenicity [54]
    2. Embryotoxicity [55]

Examples of predicting the biological activity of organic compounds

  1. The spectrum of biological activity [56]
  2. Belonging to pharmacological groups [57]

See also

  • Chemoinformatics
  • Computer chemistry
  • Pharmaceutical chemistry
  • Mathematical statistics
  • Machine learning

Literature

  1. ↑ NS Zefirov, VA Palyulin. Fragmental Approach in QSPR (Eng.) // J. Chem. Inf. Comput. Sci. : journal. - 2002. - Vol. 42 , no. 5 . - P. 1112-1122 . - DOI : 10.1021 / ci020010e .
  2. ↑ I. Baskin, A. Varnek. Building a chemical space based on fragment descriptors (Eng.) // Combinatorial Chemistry & High Throughput Screening : journal. - 2008. - Vol. 11 , no. 8 . - P. 661–668 .
  3. ↑ 1 2 3 I. Baskin, A. Varnek. "Fragment Descriptors in SAR / QSAR / QSPR Studies, Molecular Similarity Analysis and in Virtual Screening." In: Chemoinformatic Approaches to Virtual Screening , A. Varnek, A. Tropsha, eds., RCS Publishing, 2008, ISBN 978-0-85404-144-2 , P. 1-43
  4. ↑ I. I. Baskin, M. I. Skvortsova, I. V. Stankevich, N. S. Zefirov . On the basis of invariants of labeled molecular graphs (Russian) // Dokl. RAS: journal. - 1994. - T. 339 , No. 3 . - S. 346-350 .
  5. ↑ II Baskin, MI Skvortsova, IV Stankevich, NS Zefirov. On basis of invariants of labeled molecular graphs (Eng.) // J. Chem. Inf. Comput. Sci. : journal. - 1995. - Vol. 35 , no. 3 . - P. 527-531 . - DOI : 10.1021 / ci00025a021 .
  6. ↑ O.A. Raevsky. Descriptors of molecular structure in the computer design of biologically active substances (rus.) // Uspekhi Khimii : Journal. - Russian Academy of Sciences , 1999. - T. 68 , No. 6 . - S. 555-575 .
  7. ↑ O.A. Raevsky. Hydrogen bond descriptors in computer molecular design (rus.) // Ros. Chem. g. (J. Ros. Chem. Ob. Named after D.I. Mendeleev): journal. - 2006. - T. L , No. 2 . - S. 97-107 .
  8. ↑ M. Karelson, VS Lobanov, AR Katritzky. Quantum-Chemical Descriptors in QSAR / QSPR Studies (Eng.) // Chem. Rev. : journal. - 1996. - Vol. 96 , no. 3 . - P. 1027-1044 . - DOI : 10.1021 / cr950202r .
  9. ↑ RD Cramer, DE Patterson, JD Bunce. Comparative molecular field analysis (CoMFA). 1. Effect of shape on binding of steroids to carrier proteins (English) // J. Am. Chem. Soc. : journal. - 1988. - Vol. 110 , no. 18 . - P. 5959-5967 . - DOI : 10.1021 / ja00226a005 .
  10. ↑ Palm, V. A. Fundamentals of the quantitative theory of organic reactions. - 2nd, per. and additional .. - L .: Chemistry, 1977. - 360 p.
  11. ↑ F. Bonachera, B. Parent, F. Barbosa, N. Froloff, D. Horvath. Fuzzy Tricentric Pharmacophore Fingerprints. 1. Topological Fuzzy Pharmacophore Triplets and Adapted Molecular Similarity Scoring Schemes (Eng.) // J. Chem. Inf. Model. : journal. - 2006. - Vol. 46 , no. 6 . - P. 2457-2477 . - DOI : 10.1021 / ci6002416 .
  12. ↑ R. Todeschini, V. Consonni: Handbook of Molecular Descriptors . WILEY-WCH Publishers, Weinheim, 2000. ISBN 3-527-29913-0
  13. ↑ 1 2 Baskin I.I., Majidov T.I., Varnek A.A. Introduction to chemoinformatics: a training manual. Part 3. Modeling "structure-property" , Kazan: Kazan University, 2015, ISBN 978-5-00019-442-3
  14. ↑ 1 2 Baskin I.I., Majidov T.I., Varnek A.A. Introduction to chemoinformatics: a training manual. Part 4. Methods of machine learning , Kazan: Kazan University, 2016, ISBN 978-5-00019-695-3
  15. ↑ I. I. Baskin, V. A. Palyulin, N. S. Zefirov. Methodology for the search for direct correlations between the structures and properties of organic compounds using computational neural networks (Russian) // Dokl. RAS: journal. - 1993. - T. 333 , No. 2 . - S. 176-179 .
  16. ↑ II Baskin, VA Palyulin, NS Zefirov. A Neural Device for Searching Direct Correlations between Structures and Properties of Organic Compounds // J. Chem. Inf. Comput. Sci. : journal. - 1997. - Vol. 37 , no. 4 . - P. 715-721 . - DOI : 10.1021 / ci940128y .
  17. ↑ DB Kireev. ChemNet: A Novel Neural Network Based Method for Graph / Property Mapping // J. Chem. Inf. Comput. Sci. : journal. - 1995. - Vol. 35 , no. 2 . - P. 175-180 . - DOI : 10.1021 / ci00024a001 .
  18. ↑ AM Bianucci. Application of Cascade Correlation Networks for Structures to Chemistry // Applied Intelligence: journal. - 2000. - Vol. 12 , no. 1-2 . - P. 117-146 .
  19. ↑ A. Micheli, A. Sperduti, A. Starita, AM Bianucci. Analysis of the Internal Representations Developed by Neural Networks for Structures Applied to Quantitative Structure-Activity Relationship Studies of Benzodiazepines // J. Chem. Inf. Comput. Sci. : journal. - 2001. - Vol. 41 , no. 1 . - P. 202-218 . - DOI : 10.1021 / ci9903399 .
  20. ↑ O. Ivanciuc. Molecular Structure Encoding into Artificial Neural Networks Topology (Eng.) // Roumanian Chemical Quarterly Reviews: journal. - 2001. - Vol. 8 . - P. 197-220 .
  21. ↑ A. Goulon, T. Picot, A. Duprat, G. Dreyfus. Predicting activities without computing descriptors: Graph machines for QSAR (English) // SAR and QSAR in Environmental Research: journal. - 2007. - Vol. 18 , no. 1-2 . - P. 141-153 . - DOI : 10.1080 / 10629360601054313 .
  22. ↑ H. Kashima, K. Tsuda, A. Inokuchi, Marginalized Kernels Between Labeled Graphs, The 20th International Conference on Machine Learning (ICML2003), 2003. PDF
  23. ↑ H. Fröhlich, JK Wegner, A. Zell, Optimal Assignment Kernels For Attributed Molecular Graphs , The 22nd International Conference on Machine Learning (ICML 2005), Omnipress, Madison, WI, USA, 2005 , 225—232. Pdf
  24. ↑ H. Fröhlich, JK Wegner, A. Zell. Kernel Functions for Attributed Molecular Graphs - A New Similarity Based Approach To ADME Prediction in Classification and Regression (англ.) // QSAR Comb. Sci. : journal. — 2006. — Vol. 25 , no. 4 . — P. 317—326 . — DOI : 10.1002/qsar.200510135 .
  25. ↑ H. Fröhlich, JK Wegner, A. Zell, Assignment Kernels For Chemical Compounds , International Joint Conference on Neural Networks 2005 (IJCNN'05), 2005 , 913—918. CiteSeer
  26. ↑ P. Mahe, L. Ralaivola, V. Stoven, J. Vert. The pharmacophore kernel for virtual screening with support vector machines (англ.) // J. Chem. Inf. Model. : journal. — 2006. — Vol. 46 , no. 5 . — P. 2003—2014 . — DOI : 10.1021/ci060138m .
  27. ↑ 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 AR Katritzky, U. Maran, VS Lobanov, M. Karelson. Structurally Diverse Quantitative Structure-Property Relationship Correlations of Technologically Relevant Physical Properties (англ.) // J. Chem. Inf. Comput. Sci. : journal. — 2000. — Vol. 40 , no. 1 . — P. 1—18 . — DOI : 10.1021/ci9903206 .
  28. ↑ 1 2 3 4 Н. В. Артеменко, И. И. Баскин, В. А. Палюлин, Н. С. Зефиров. Прогнозирование физических свойств органических соединений при помощи искусственных нейронных сетей в рамках подструктурного подхода (рус.) // Докл. АН : журнал. — 2001. — Т. 381 , № 2 . — С. 203—206 .
  29. ↑ 1 2 3 Н. В. Артеменко, И. И. Баскин, В. А. Палюлин, Н. С. Зефиров. Искусственные нейронные сети и фрагментный подход в прогнозировании физико-химических свойств органических соединений (рус.) // Известия РАН, Серия химическая : журнал. — 2003. — № 1 . — С. 19—28 .
  30. ↑ Н. И. Жохова, И. И. Баскин, В. А. Палюлин, А. Н. Зефиров, Н. С. Зефиров. Фрагментные дескрипторы в QSPR: применение для расчета поляризуемости молекул (рус.) // Известия РАН, Серия химическая : журнал. — 2003. — № 5 . — С. 1005—1009 .
  31. ↑ Н. И. Жохова, И. И. Баскин, В. А. Палюлин, А. Н. Зефиров, Н. С. Зефиров. Фрагментные дескрипторы в QSPR: применение для расчета магнитной восприимчивости (рус.) // Журнал структурной химии : журнал. — 2004. — Т. 45 , № 4 . — С. 660—669 .
  32. ↑ Н. И. Жохова, И. И. Баскин, В. А. Палюлин, А. Н. Зефиров, Н. С. Зефиров. Расчет энтальпии сублимации методом QSPR с применением фрагментного подхода (рус.) // Журнал прикладной химии : журнал. — 2003. — Т. 76 , № 12 . — С. 1966—1970 .
  33. ↑ 1 2 3 Н. М. Гальберштам, И. И. Баскин, В. А. Палюлин, Н. С. Зефиров. Построение нейросетевых зависимостей структура-условия-свойство. Моделирование физико-химических свойств углеводородов (рус.) // Докл. АН : журнал. — 2002. — Т. 384 , № 2 . — С. 202—205 .
  34. ↑ И. И. Баскин, А. О. Айт, Н. М. Гальберштам, В. А. Палюлин, Н. С. Зефиров. Применение методологии искусственных нейронных сетей для прогнозирования свойств сложных молекулярных систем. Предсказание положения длинноволновой полосы поглощения симметричных цианиновых красителей (рус.) // Докл. АН : журнал. — 1997. — Т. 357 , № 1 . — С. 57—59 .
  35. ↑ Y. Binev, J. Aires-de-Sousa. Structure-Based Predictions of 1 H NMR Chemical Shifts Using Feed-Forward Neural Networks (англ.) // J. Chem. Inf. Comput. Sci. : journal. — 2004. — Vol. 44 , no. 3 . — P. 940—945 . — DOI : 10.1021/ci034228s .
  36. ↑ JP Doucet, A. Panaye, E. Feuilleaubois, P. Ladd. Neural Networks and 13 C NMR Shift Prediction (англ.) // J. Chem. Inf. Comput. Sci. : journal. — 1993. — Vol. 33 , no. 3 . — P. 320—324 . — DOI : 10.1021/ci00013a007 .
  37. ↑ GMJ West. Predicting Phosphorus NMR Shifts Using Neural Networks (англ.) // J. Chem. Inf. Comput. Sci. : journal. — 1993. — Vol. 33 , no. 4 . — P. 577—589 . — DOI : 10.1021/ci00014a009 .
  38. ↑ Н. И. Жохова, И. И. Баскин, В. А. Палюлин, А. Н. Зефиров, Н. С. Зефиров. Фрагментные дескрипторы в QSAR: применение для расчета температуры вспышки (рус.) // Известия РАН, Серия химическая : журнал. — 2003. — № 9 . — С. 1787—1793 .
  39. ↑ А. В. Сидорова, И. И. Баскин, В. А. Палюлин, Д. Е. Петелин, Н. С. Зефиров. Исследование зависимостей между структурой и октановыми числами углеводородов (рус.) // Докл. АН : журнал. — 1996. — Т. 350 , № 5 . — С. 642—646 .
  40. ↑ А. А. Иванова, И. И. Баскин, В. А. Палюлин, Н. С. Зефиров. Оценка значений констант ионизации для различных классов органических соединений с использованием фрагментного подхода к поиску зависимостей «структура-свойство» (рус.) // Докл. АН : журнал. — 2007. — Т. 413 , № 6 . — С. 766—770 .
  41. ↑ D. Butina, JMR Gola. Modeling Aqueous Solubility (англ.) // J. Chem. Inf. Comput. Sci. : journal. - 2003. - Vol. 43 , no. 3 . — P. 837—841 . — DOI : 10.1021/ci020279y .
  42. ↑ IV Tetko, V. Yu. Tanchuk, AEP Villa. Prediction of n-Octanol/Water Partition Coefficients from PHYSPROP Database Using Artificial Neural Networks and E-State Indices (англ.) // J. Chem. Inf. Comput. Sci. : journal. — 2001. — Vol. 41 , no. 5 . — P. 1407—1421 . — DOI : 10.1021/ci010368v .
  43. ↑ В. С. Бугрин, М. Ю. Козлов, И. И. Баскин, Н. С. Мелик-Нубаров. Межмолекулярные взаимодействия, определяющие солюбилизацию в мицеллах полиалкиленоксидных поверхностно активных веществ (рус.) // Высокомолекулярные соединения, Серия А : журнал. — 2007. — Т. 49 , № 4 . — С. 701—712 .
  44. ↑ А. А. Кравцов, П. В. Карпов, И. И. Баскин, В. А. Палюлин, Н. С. Зефиров. «Бимолекулярный» QSPR: Оценка свободной энергии сольватации органических молекул в различных растворителях (рус.) // Докл. АН : журнал. — 2007. — Т. 414 , № 3 . — С. ??? .
  45. ↑ NM Halberstam, II Baskin, VA Palyulin, NS Zefirov. Quantitative Structure - Conditions - Property Relationships Studies. The Network Modelling of the Neural the Acid Hydrolysis of Esters (Eng.) : Journal. - 2002. - Vol. 12 , no. 6 . - P. 185-186 .
  46. ↑ N.I. Zhokhov, E.V. Bobkov, I.I. Baskin, V.A. Palyulin, A.N. Zefirov, N.S. Zefirov. Calculation of the stability of complexes of organic compounds with β-cyclodextrin using the QSPR method (Russian) // Vestnik MGU, ser. 2, Chemistry: Journal. - 2007. - T. 48 , No. 5 . - S. 329-332 .
  47. ↑ N.I. Zhokhov, I.I. Baskin, V.A. Palyulin, A.N. Zefirov, N.S. Zefirov. The study of the affinity of dyes for cellulose fiber as part of a fragmented approach in QSPR (Russian) // Journal of Applied Chemistry: Journal. - 2005. - T. 78 , No. 6 . - S. 1034-1037 .
  48. ↑ IV Tetko, VP Solov'ev, AV Antonov, X. Yao, JP Doucet, B. Fan, F. Hoonakker, D. Fourches, P. Jost, N. Lachiche, A. Varnek. Benchmarking of Linear and Nonlinear Approaches for Quantitative Structure-Property Relationship Studies of Metal Complexation with Ionophores // J. Chem. Inf. Model. : journal. - 2006. - Vol. 46 , no. 2 . - P. 808-819 . - DOI : 10.1021 / ci0504216 .
  49. ↑ MP Gonzalez, AM Helguera, HG Diaz. A TOPS-MODE approach to predict permeability coefficients // Polymer: journal. - 2004. - Vol. 45 . - P. 2073-2079 .
  50. ↑ Varnek, A .; Kireeva, N .; Tetko, IV; Baskin, II; Solov'ev, VP Exhaustive QSPR Studies of Large Diverse Set of Ionic Liquids: How Accurately Can We Predict Melting Points? (English) // J. Chem. Inf. Model. : journal. - 2007. - Vol. 47 . - P. 1111-1122 .
  51. ↑ AR Katritzky, M. Kuanar, S. Slavov, DA Dobchev, DC Fara, M. Karelson, WE Acree, Jr., VP Solov'ev, A. Varnek. Correlation of blood-brain penetration using structural descriptors (Eng.) // Bioorg. Med. Chem. : journal. - 2006. - Vol. 14 . - P. 4888-4917 . - DOI : 10.1016 / j.bmc.2006.03.012 .
  52. ↑ AR Katritzky, DA Dobchev, DC Fara, E. Hur, K. Tamm, L. Kurunczi, M. Karelson, A. Varnek, VP Solov'ev. Skin permeation rate as a function of chemical structure (Eng.) // J. Med. Chem. : journal. - 2006. - Vol. 49 , no. 11 . - P. 3305–3314 . - DOI : 10.1021 / jm051031d .
  53. ↑ Yu. Borodina, A., Rudik, D. Filimonov, N. Kharchevnikova, A. Dmitriev, V. Blinova, V. Poroikov. New Statistical Approach to Predicting Aromatic Hydroxylation Sites. Comparison with Model-Based Approaches // J. Chem. Inf. Comput. Sci. : journal. - 2004. - Vol. 44 , no. 6 . - P. 1998-2009 . - DOI : 10.1021 / ci049834h .
  54. ↑ AA Lagunin, JC Dearden, DA Filimonov, VV Poroikov. Computer-aided rodent carcinogenicity prediction (neopr.) // Mutation Research . - Elsevier , 2005. - T. 586 , No. 2 . - S. 138-146 . - PMID 16112600 .
  55. ↑ I. I. Baskin, G. A. Buznikov, A. S. Kabankin, M. A. Landau, L. A. Leksina, A. A. Ordukhanyan, V. A. Palyulin, N. S. Zefirov. Computer study of the relationship between embryotoxicity and the structures of synthetic analogues of biogenic amines (rus.) // Bulletin of the Russian Academy of Sciences, biological series: journal. - 1997. - No. 4 . - S. 407-413 .
  56. ↑ D.A. Filimonov and V.V. Poroykov. Forecast of the spectrum of biological activity of organic compounds (rus.) // Ros. Chem. g. (J. Ros. Chem. Ob. Named after D.I. Mendeleev): journal. - 2006. - T. L , No. 2 . - S. 66-75 .
  57. ↑ E.P. Kondratovich, N.I. Zhokhov, I.I. Baskin, V.A. Palyulin, N.S. Zefirov. Fragment descriptors in the study of structure-activity ratios: application for predicting the membership of organic compounds in pharmacological groups using the support vector method (Russian) // Izvestiya Akademii Nauk. Chemical series: journal. - 2009 .-- T. 4 . - S. 641-647 .

Links

  • International Society of Chemoinformatics and QSAR
  • UK Chemoinformatics & QSAR Group (link unavailable)
  • QSAR World - Global QSAR Network Resource
  • QSAR History - ( PDF )
Source - https://ru.wikipedia.org/w/index.php?title=QSAR&oldid=100958114


More articles:

  • Coordination of Movement
  • Asaba (area)
  • pwdump
  • Vladimir (Kantaryan)
  • Pirates Online
  • Detenamo, Itte
  • Pargel, Anka
  • Crofts, Freeman Wills
  • Berkeley, Anthony
  • Receptor, Vladimir Emanuilovich

All articles

Clever Geek | 2019