Robust QSAR models using Bayesian regularized neural networks.

F R Burden; D A Winkler

doi:10.1021/jm980697n

Home/Archive/PubMed

PubMed · 10447964

Robust QSAR models using Bayesian regularized neural networks.

F R Burden·D A Winkler

View original source ↗Export metadata ↓

Abstract

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.

Explore related subjects

Drug Design Neural Networks, Computer Regression Analysis Structure-Activity Relationship Comparative Study Journal Article

Keep this discovery

Explore connections, maps & timelines

BibTeX RIS

F R Burden, D A Winkler. 1999-08-12. Robust QSAR models using Bayesian regularized neural networks.. https://doi.org/10.1021/jm980697n

Cite the original work for its findings. Save a collection to share your selection of sources.

KEEP EXPLORING

Related citations

PubMed2002

Rational design of 4,5-disubstituted-5,7-dihydro-pyrrolo[2,3-d]pyrimidin-6-ones as a novel class of inhibitors of epidermal growth factor receptor (EGF-R) and Her2(p185(erbB)) tyrosine kinases.

A novel class of 4,5-disubstituted-5,7-dihydro-pyrrolo[2,3-d]pyrimidin-6-ones has been discovered as potent and selective inhibitors of the EGF-R tyrosine kinase family. These compounds selectively inhibit EGF-R kinase activity at low nanomolar concentration and tyrosine autophosphorylation in cells expressing EGF-R or Her2 (p185(erbB)). Structure-activity relationships (SARs) for this class of compounds are presented.

Drug Design↗

PubMed2002

Molecular design of antifungal agents.

In a rational approach to the design of antifungal agents against Saccharomyces cerevisiae, a series of alkyl gallates (3,4,5-trihydroxybenzoates) were synthesized and assayed. Nonyl gallate (1) was found to be the most effective with a minimum fungicidal concentration (MFC) of 12.5 microg/mL (42 microM), followed by octyl gallate (2) with an MFC of 25 microg/mL (89 microM). These MFCs are little influenced by pH values. A time-kill curve study indicates that nonyl gallate exhibits fungicidal activity against S. cerevisiae at any growing stage. The antifungal activity of nonyl gallate is due primarily to its ability to act as a nonionic surface-active agent (surfactant). The length of the alkyl group is not a major contributor but plays a role in eliciting the activity to a large extent. As far as alkyl gallates are concerned, their antimicrobial spectra and potency depend largely on the hydrophobic portion of the molecules.

Drug Design↗

PubMed2002

Drugs, leads, and drug-likeness: an analysis of some recently launched drugs.

An analysis of the origins of recently launched drugs reveals that most were derived by modification of known drug structures or from lead structures obtained from the scientific literature. High-throughput screening did not have a significant impact on the derivation of these drugs. The drug structures are very closely related to their leads.

Drug Design↗