Comparative Study of Factor Xa Inhibitors Using Molecular Docking/SVM/HQSAR/3D‐QSAR Methods

Sun, Jing; Chen, Hai Feng; Xia, Hai; Yao, Jian; Fan, Bo

doi:10.1002/qsar.200530115

Cited by 8 publications

(5 citation statements)

References 43 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…There have been previous applications of kernel methods, in particular in the form of support vector machines (SVMs), to predictive problems in chemistry. − Most of the previous work, however, focuses on binary classification problems (e.g., toxic/nontoxic) rather than regression problems, where the goal is to predict a numerical value associated with a particular property of a molecule (e.g., melting point). With the exception of the NCI data sets used in Swamidass et al., most of the previous applications are based on very small data sets containing at most a few hundred examples, and often much less.…”

Section: Introductionmentioning

confidence: 99%

One- to Four-Dimensional Kernels for Virtual Screening and the Prediction of Physical, Chemical, and Biological Properties

Azencott

Ksikes

Swamidass

et al. 2007

J. Chem. Inf. Model.

View full text Add to dashboard Cite

Many chemoinformatics applications, including high-throughput virtual screening, benefit from being able to rapidly predict the physical, chemical, and biological properties of small molecules to screen large repositories and identify suitable candidates. When training sets are available, machine learning methods provide an effective alternative to ab initio methods for these predictions. Here, we leverage rich molecular representations including 1D SMILES strings, 2D graphs of bonds, and 3D coordinates to derive efficient machine learning kernels to address regression problems. We further expand the library of available spectral kernels for small molecules developed for classification problems to include 2.5D surface and 3D kernels using Delaunay tetrahedrization and other techniques from computational geometry, 3D pharmacophore kernels, and 3.5D or 4D kernels capable of taking into account multiple molecular configurations, such as conformers. The kernels are comprehensively tested using cross-validation and redundancy-reduction methods on regression problems using several available data sets to predict boiling points, melting points, aqueous solubility, octanol/water partition coefficients, and biological activity with state-of-the art results. When sufficient training data are available, 2D spectral kernels in general tend to yield the best and most robust results, better than state-of-the art. On data sets containing thousands of molecules, the kernels achieve a squared correlation coefficient of 0.91 for aqueous solubility prediction and 0.94 for octanol/water partition coefficient prediction. Averaging over conformations improves the performance of kernels based on the three-dimensional structure of molecules, especially on challenging data sets. Kernel predictors for aqueous solubility (kSOL), LogP (kLOGP), and melting point (kMELT) are available over the Web through: http:// cdb.ics.uci.edu.

show abstract

Section: Introductionmentioning

confidence: 99%

One- to Four-Dimensional Kernels for Virtual Screening and the Prediction of Physical, Chemical, and Biological Properties

Azencott

Ksikes

Swamidass

et al. 2007

J. Chem. Inf. Model.

View full text Add to dashboard Cite

show abstract

“…Blue-colored regions show areas where electropositive charged groups enhance MK-2 inhibitory activity, while red regions represent where electronegative charged groups improve the activity. (12), (15) and (18) (with a substituents at the 3-position) have more inhibitory activity than compounds (11), (14) and (17). At the 2-position, there is a relatively large blue region and a small yellow region.…”

Section: Resultsmentioning

confidence: 95%

3D-QSAR Analysis on Pyrrolopyridine Analogs as Mitogen-Activated Protein Kinase-Activated Protein Kinase 2 (MK-2) Inhibitors

Yang¹,

Sun²,

Chen³

2007

LDDD

View full text Add to dashboard Cite

Mitogen-activated protein kinase-activated protein kinase 2 (MK-2) plays an important role in treatment of inflammatory disease such as rheumatoid arthritis. CoMFA was conducted on thirty-one analogs displaying variable inhibition of MK-2 to determine the structural requirements for potency in inhibiting MK-2. The resulting model exhibited good q 2 and r 2 values up to 0.549 and 0.973 for CoMFA, the standard error of estimation was 0.098. The contributions from the steric and electrostatic fields were 0.586 and 0.414 respectively. The 3D-QSAR model should be very useful for design of novel MK-2 inhibitors.

show abstract

“…3, illustrating how it identified a pair of Factor Xa inhibitors from the literature. 17 A set of 400 functional groups was selected from the most common functional groups used in Pfizer drug discovery programs. The SWAP methodology was applied to these groups to generate a secondary database (the SWAP database) containing matched pairs of compounds differing structurally by a single functional group change from one of these groups.…”

Section: Pairwise Analysismentioning

confidence: 99%

SAR mining and its application to the design of TRPA1 antagonists

et al. 2012

View full text Add to dashboard Cite

Given the large amounts of screening data now available, empirical methods derived from matchedmolecular pairs are being used as a means for suggesting bioisosteric replacements to the medicinal chemist. The pairwise analysis of compounds has been extended to the pairwise analysis of series to bring further context to these suggestions. A validation dataset derived from recent literature has been used to demonstrate that, given a series of active compounds, this approach would be expected to predict a more potent compound, if it exists, in around 46% of cases. The approach has been successfully applied to a series of TRPA1 antagonists.

show abstract

Comparative Study of Factor Xa Inhibitors Using Molecular Docking/SVM/HQSAR/3D‐QSAR Methods

Cited by 8 publications

References 43 publications

One- to Four-Dimensional Kernels for Virtual Screening and the Prediction of Physical, Chemical, and Biological Properties

One- to Four-Dimensional Kernels for Virtual Screening and the Prediction of Physical, Chemical, and Biological Properties

3D-QSAR Analysis on Pyrrolopyridine Analogs as Mitogen-Activated Protein Kinase-Activated Protein Kinase 2 (MK-2) Inhibitors

SAR mining and its application to the design of TRPA1 antagonists

Contact Info

Product

Resources

About