A fast virtual screening filter for cytochrome P450 3A4 inhibition liability of compound libraries

Zuegge, Jochen; Fechner, Uli; Roche, Olivier; Parrott, Neil; Engkvist, Ola; Schneider, Gisbert

doi:10.1002/1521-3838(200208)21:3<249::aid-qsar249>3.0.co;2-s

Cited by 46 publications

(40 citation statements)

References 17 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…The descriptor set consists of 146 2D-descriptors from MOE, 17 the 120 Ghose and Crippen descriptors (GC), 15 and an additional set of 63 topological, electronic, count, and structural descriptors. 16 Again, a lower-dimensional variant of this data set was compiled using only GC descriptors as input variables which is denoted as CYP GC. Table 1 and Figure 3 show the cross-validated classification rates and cross-validated Matthew's correlation coefficients (CC) for two data sets and the six different processing schemes described here:…”

Section: Data Sets the Proposed Methods Has Been Applied To Two Data mentioning

confidence: 99%

Ensemble Methods for Classification in Cheminformatics

Merkwirth

Mauser

Schulz‐Gasch

et al. 2004

J. Chem. Inf. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

We describe the application of ensemble methods to binary classification problems on two pharmaceutical compound data sets. Several variants of single and ensembles models of k-nearest neighbors classifiers, support vector machines (SVMs), and single ridge regression models are compared. All methods exhibit robust classification even when more features are given than observations. On two data sets dealing with specific properties of drug-like substances (cytochrome P450 inhibition and "Frequent Hitters", i.e., unspecific protein inhibition), we achieve classification rates above 90%. We are able to reduce the cross-validated misclassification rate for the Frequent Hitters problem by a factor of 2 compared to previous results obtained for the same data set with different modeling techniques.

show abstract

Section: Data Sets the Proposed Methods Has Been Applied To Two Data mentioning

confidence: 99%

Ensemble Methods for Classification in Cheminformatics

Merkwirth

Mauser

Schulz‐Gasch

et al. 2004

J. Chem. Inf. Comput. Sci.

Self Cite

View full text Add to dashboard Cite

show abstract

“…The data acquired from such studies have been used later to develop in silico structureactivity relationship models of CYP3A4 inhibition (Table 1) [8][9][10][11][12][13][14][15][16][17], which can serve as virtual screening tools in evaluating the possibility for new compounds to inhibit CYP3A4. Such approach is very attractive because in silico models may be applied in early stages of drug discovery at a very small cost.…”

Section: Introductionmentioning

confidence: 99%

Trainable structure–activity relationship model for virtual screening of CYP3A4 inhibition

Didžiapetris¹,

Dapkūnas

Sazonovas³

et al. 2010

J Comput Aided Mol Des

View full text Add to dashboard Cite

A new structure-activity relationship model predicting the probability for a compound to inhibit human cytochrome P450 3A4 has been developed using data for >800 compounds from various literature sources and tested on PubChem screening data. Novel GALAS (Global, Adjusted Locally According to Similarity) modeling methodology has been used, which is a combination of baseline global QSAR model and local similarity based corrections. GALAS modeling method allows forecasting the reliability of prediction thus defining the model applicability domain. For compounds within this domain the statistical results of the final model approach the data consistency between experimental data from literature and PubChem datasets with the overall accuracy of 89%. However, the original model is applicable only for less than a half of PubChem database. Since the similarity correction procedure of GALAS modeling method allows straightforward model training, the possibility to expand the applicability domain has been investigated. Experimental data from PubChem dataset served as an example of in-house high-throughput screening data. The model successfully adapted itself to both data classified using the same and different IC₅₀ threshold compared with the training set. In addition, adjustment of the CYP3A4 inhibition model to compounds with a novel chemical scaffold has been demonstrated. The reported GALAS model is proposed as a useful tool for virtual screening of compounds for possible drug-drug interactions even prior to the actual synthesis.

show abstract

“…Although strategies have been developed to increase throughput by techniques such as sample pooling, multiple assay compression, high-speed LC/MS (Bu et al, 2001;Zhang et al, 2002), single-concentration IC 50 projection (Gao et al, 2002), and virtual screening (Rao et al, 2000;Zuegge et al, 2002), the analytical approach still remains a challenge that limits sample throughput.…”

mentioning

confidence: 99%

In Vitro Drug Interactions of Cytochrome P450: An Evaluation of Fluorogenic to Conventional Substrates

Cohen¹,

Remley²,

Raunig³

et al. 2003

Drug Metab Dispos

124

109

View full text Add to dashboard Cite

This article is available online at http://dmd.aspetjournals.org ABSTRACT:Clinically observed drug interactions with cytochrome P450 (P450) enzymes have increased the need to assess drug interactions of new chemical entities early in the discovery process. To meet this need, fluorogenic substrates have been commercialized. However, only limited evaluations of their utility and comparisons to drug probes have been reported. This study examines the correlation between IC 50 values obtained with fluorogenic and conventional drug probes for structurally diverse inhibitors of the five major human P450 isoforms. In general, correlations are weak, with significant numbers of compounds being missed as inhibitors by either probe. For P450s 1A2, 2C9, and 2C19, correlation coefficients were above 0.6 with slopes that ranged from 1.5 to 4.2.However, for P450s 1A2 and 2C9, about 20% of compounds were not included because an IC 50 value could not be determined with one of the two probes. CYP 2C19 had the highest correlation (correlation coefficient 0.84), with a slope of 2.0 and less than 5% of compounds excluded. CYP 2D6 showed a good correlation for IC 50 values less than 10 M. However, at higher IC 50 values, a high degree of scatter was observed. CYP 3A4 had the weakest correlation, and a large number of compounds were excluded with the fluorogenic probe. Overall, the study shows the care needed when selecting fluorogenic probes and the caution needed when results with fluorogenic probes are used to drive structure-activity relationships with respect to drug interactions.

show abstract

A fast virtual screening filter for cytochrome P450 3A4 inhibition liability of compound libraries

Cited by 46 publications

References 17 publications

Ensemble Methods for Classification in Cheminformatics

Ensemble Methods for Classification in Cheminformatics

Trainable structure–activity relationship model for virtual screening of CYP3A4 inhibition

In Vitro Drug Interactions of Cytochrome P450: An Evaluation of Fluorogenic to Conventional Substrates

Contact Info

Product

Resources

About