Classification and regression tree analysis for molecular descriptor selection and retention prediction in chromatographic quantitative structure–retention relationship studies

Put, R.; Perrin, Catherine; Questier, Frederik; Coomans, Danny; Massart, D.L.; Heyden, Yvan Vander

doi:10.1016/s0021-9673(03)00004-9

Cited by 80 publications

(44 citation statements)

References 27 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…This maximal tree will usually contain too many leaves and will overfit the learning data set, which will cause poor predictive abilities for new sample. 10 Therefore, the selection of an optimal tree with a good compromise between model fit and predictive properties is required. Thus, in general, CART analysis consists of three steps: (i) the maximal-tree building, (ii) the tree "pruning", which consists of the cutting-off of nodes to generate a sequence of simpler trees, and (iii) the optimal tree selection.…”

Section: Theorymentioning

confidence: 99%

“…10 CART is extensively used for modeling and classification in several areas, such as medical diagnosis and prognosis, [11][12][13] ecology, 14 agriculture 15 and chemistry. 10,[16][17] A very interesting advantage of CART is the possibility to deal with large numbers of both categorical and numerical variables. Another advantage is that no assumption about the underlying distribution of the predictor variables is required (even categorical variables can be used).…”

Section: Introductionmentioning

confidence: 99%

“…It means that the model has been overfitted. 10,18 Such trees are often difficult to interpret and their predictive ability for new observations is poor in general, since they tend to fit also the noise in the data. The selection of a smaller tree, derived from the maximal is then necessary for predictive purposes.…”

mentioning

confidence: 99%

“…In cross validation, some samples are randomly drawn from the data set, to test the tree, which is built with the rest of the data. 10,17 For a ten-fold cross validation, the original data set is divided into ten equal pairs (test sets), each containing a similar distribution for the response variable. A tree is then built using 90% of the observations (learning set), while the remaining 10% (test set) is used to test the tree.…”

mentioning

confidence: 99%

See 3 more Smart Citations

Classification and Regression Tree Analysis for Molecular Descriptor Selection and Binding Affinities Prediction of Imidazobenzodiazepines in Quantitative Structure-Activity Relationship Studies

Atabati¹,

Zarei²,

Abdinasab³

2009

Bulletin of the Korean Chemical Society

View full text Add to dashboard Cite

The use of the classification and regression tree (CART) methodology was studied in a quantitative structure-activity relationship (QSAR) context on a data set consisting of the binding affinities of 39 imidazobenzodiazepines for the α1 benzodiazepine receptor. The 3-D structures of these compounds were optimized using HyperChem software with semiempirical AM1 optimization method. After optimization a set of 1481 zero-to three-dimentional descriptors was calculated for each molecule in the data set. The response (dependent variable) in the tree model consisted of the binding affinities of drugs. Three descriptors (two topological and one 3D-Morse descriptors) were applied in the final tree structure to describe the binding affinities. The mean relative error percent for the data set is 3.20%, compared with a previous model with mean relative error percent of 6.63%. To evaluate the predictive power of CART cross validation method was also performed.

show abstract

Section: Theorymentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

mentioning

confidence: 99%

mentioning

confidence: 99%

See 2 more Smart Citations

Classification and Regression Tree Analysis for Molecular Descriptor Selection and Binding Affinities Prediction of Imidazobenzodiazepines in Quantitative Structure-Activity Relationship Studies

Atabati¹,

Zarei²,

Abdinasab³

2009

Bulletin of the Korean Chemical Society

View full text Add to dashboard Cite

show abstract

“…CART itself was already used successfully in quantitative structure-retention relationships by Put et al [12] and in quantitative structure-enantioselectivity relationships by Caetano et al [13]. In a QSAR-context, our group [14] and Bai et al [15] showed that CART could be used to predict the gastrointestinal absorption of drugs.…”

Section: Introductionmentioning

confidence: 97%

Evaluation of boosted regression trees (BRTs) and two‐step BRT procedures to model and predict blood‐brain barrier passage

Deconinck

Zhang

Coomans

et al. 2007

Journal of Chemometrics

View full text Add to dashboard Cite

Two new approaches, boosted regression trees (BRTs) and two-step BRT, were evaluated for modelling and predicting the blood-brain barrier (BBB) passage of drugs. Classification and regression trees (CART) were used as a base learner in BRT. In two-step BRT, a linear model (stepwise multiple linear regression (MLR) or partial least squares (PLS)) was built first, then BRT was applied to model the residuals of the linear model and both models were added. Both approaches were compared with the CART, MLR and PLS models. It was observed that BRT could improve the descriptive and predictive abilities compared to a single CART and that the stepwise MLR-BRT results in slightly improved descriptive and predictive properties compared to the MLR model. The combination of PLS and BRT did not result in an improvement, compared to the individual PLS model. The best models were obtained with stepwise MLR-BRT and PLS. It was shown that the combination of linear models with BRT is an approach that has potential and can be considered for future QSAR modelling.

show abstract

Recent Advances in Development, Validation, and Exploitation ofQSARModels

Tropsha

2010

Burger's Medicinal Chemistry and Drug Discovery

View full text Add to dashboard Cite

Counting more than 45 years of intensive development and applications, QSAR modeling remains one of the major computational approaches for drug discovery and environmental chemical risk assessment. Recent developments in the field reflect the burgeoning growth of bioactivity and chemical databases. This growth provides a never‐ending impetus for the development of a growing body of QSAR models and their application for virtual screening to prioritize chemicals for experimental investigations. This chapter reviews recent and emerging trends in the QSAR modeling field with the emphasis on new approaches to chemical and biological data curation, model validation, virtual screening, and potential application of QSAR models by federal agencies for regulatory decision making.

show abstract

Classification and regression tree analysis for molecular descriptor selection and retention prediction in chromatographic quantitative structure–retention relationship studies

Cited by 80 publications

References 27 publications

Classification and Regression Tree Analysis for Molecular Descriptor Selection and Binding Affinities Prediction of Imidazobenzodiazepines in Quantitative Structure-Activity Relationship Studies

Classification and Regression Tree Analysis for Molecular Descriptor Selection and Binding Affinities Prediction of Imidazobenzodiazepines in Quantitative Structure-Activity Relationship Studies

Evaluation of boosted regression trees (BRTs) and two‐step BRT procedures to model and predict blood‐brain barrier passage

Recent Advances in Development, Validation, and Exploitation ofQSARModels

Contact Info

Product

Resources

About