QSAR models for tyrosinase inhibitory activity description applying modern statistical classification techniques: A comparative study

Le-Thi-Thu, Huong; Cardoso, Gladys Casas; Casañola-Martín, Gerardo M.; Marrero-Ponce, Yovani; Puris, Amilkar; Torrens, Francisco; Rescigno, Antonio; Abad, Concepción

doi:10.1016/j.chemolab.2010.08.016

Cited by 8 publications

(6 citation statements)

References 34 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…In this section, the behaviors of ML-based QSAR models obtained in the present report, as well as the models achieved in our previous report (LDA, QDA, BLR, CT) [8] were compared.…”

Section: Results Of Scpsmentioning

confidence: 98%

“…Finally, the different ANNs were developed on the attribute subsets obtained by the filters and wrappers. The attribute sets selected by LDA, QSA, BLR and CT in [8] were also attempted. Likewise, for each ML technique, various models were developed.…”

Section: Prediction Of Tyrosinase Inhibitory Activitymentioning

confidence: 99%

“…[13] At each step, this option includes in the model set the feature whose addition to the model results in the smallest error (computed as the error rate) and continues until the specified stopping condition is met. [16] Finally, for each technique we also attempted the attribute subsets selected by statistical techniques [LDA, quadratic discriminant analysis (QDA), binary logistic regression (BLR) and classification tree (CT)] (results of of a previous study [8] ).…”

Section: Attribute Selectionmentioning

confidence: 99%

“…The fingerprints so-called atom-based non-stochastic and stochastic quadratic indices calculated in [8] by program TO-MOCOMD-CARDD [10] are used for the present report. This method codifies molecular structures by means of mathematical quadratic forms.…”

Section: Molecular Descriptorsmentioning

confidence: 99%

“…[7] The LDA procedure generates a discriminant function (or, for more than two groups, a set of discriminant functions) based on linear combinations of the predictor variables that provide the best discrimination between the groups. In an early publication, various modern statistical methods were applied in a comparative study to describe tyrosinase inhibitory activity [8] and they showed better performance than traditional LDA. However, ML approaches, such as support vector machine (SVM), artificial neural network (ANN), Bayesian networks (BN) and k-nearest neighbor (k-NN), are known to enable the analysis of more complex, nonlinear relationships and in general, offer greater performance than their statistical forebears.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

A Comparative Study of Nonlinear Machine Learning for the “In Silico” Depiction of Tyrosinase Inhibitory Activity from Molecular Structure

Le-Thi-Thu

Marrero-Ponce

Casañola-Martín

et al. 2011

Molecular Informatics

Self Cite

View full text Add to dashboard Cite

In the preset report, for the first time, support vector machine (SVM), artificial neural network (ANN), Bayesian networks (BNs), k-nearest neighbor (k-NN) are applied and compared on two "in-house" datasets to describe the tyrosinase inhibitory activity from the molecular structure. The data set Data I is used for the identification of tyrosinase inhibitors (TIs) including 701 active and 728 inactive compounds. Data II consists of active chemicals for potency estimation of TIs. The 2D TOMOCOMD-CARDD atom-based quadratic indices are used as molecular descriptors. The derived models show rather encouraging results with the areas under the Receiver Operating Characteristic (AURC) curve in the test set above 0.943 and 0.846 for the Data I and Data II, respectively. Multiple comparison tests are carried out to compare the performance of the models and reveal the improvement of machine learning (ML) techniques with respect to statistical ones (see Chemometr. Intell. Lab. Syst. 2010, 104, 249). In some cases, these ameliorations are statistically significant. The tests also demostrate that k-NN, despite being a rather simple approach, presents the best behavior in both data. The obtained results suggest that the ML-based models could help to improve the virtual screening procedures and the confluence of these different techniques can increase the practicality of data mining procedures of chemical databases for the discovery of novel TIs as possible depigmenting agents.

show abstract

“…In this section, the behaviors of ML-based QSAR models obtained in the present report, as well as the models achieved in our previous report (LDA, QDA, BLR, CT) [8] were compared.…”

Section: Results Of Scpsmentioning

confidence: 98%

Section: Prediction Of Tyrosinase Inhibitory Activitymentioning

confidence: 99%

Section: Attribute Selectionmentioning

confidence: 99%

Section: Molecular Descriptorsmentioning

confidence: 99%

Section: Introductionmentioning

confidence: 99%

See 3 more Smart Citations

A Comparative Study of Nonlinear Machine Learning for the “In Silico” Depiction of Tyrosinase Inhibitory Activity from Molecular Structure

Le-Thi-Thu

Marrero-Ponce

Casañola-Martín

et al. 2011

Molecular Informatics

Self Cite

View full text Add to dashboard Cite

show abstract

Atom based linear index descriptors in QSAR-machine learning classifiers for the prediction of ubiquitin-proteasome pathway activity

et al. 2018

View full text Add to dashboard Cite

Exploring different strategies for imbalanced ADME data problem: case study on Caco-2 permeability modeling

et al. 2015

View full text Add to dashboard Cite

In many absorption, distribution, metabolism, and excretion (ADME) modeling problems, imbalanced data could negatively affect classification performance of machine learning algorithms. Solutions for handling imbalanced dataset have been proposed, but their application for ADME modeling tasks is underexplored. In this paper, various strategies including cost-sensitive learning and resampling methods were studied to tackle the moderate imbalance problem of a large Caco-2 cell permeability database. Simple physicochemical molecular descriptors were utilized for data modeling. Support vector machine classifiers were constructed and compared using multiple comparison tests. Results showed that the models developed on the basis of resampling strategies displayed better performance than the cost-sensitive classification models, especially in the case of oversampling data where misclassification rates for minority class have values of 0.11 and 0.14 for training and test set, respectively. A consensus model with enhanced applicability domain was subsequently constructed and showed improved performance. This model was used to predict a set of randomly selected high-permeability reference drugs according to the biopharmaceutics classification system. Overall, this study provides a comparison of numerous rebalancing strategies and displays the effectiveness of oversampling methods to deal with imbalanced permeability data problems.

show abstract

QSAR models for tyrosinase inhibitory activity description applying modern statistical classification techniques: A comparative study

Cited by 8 publications

References 34 publications

A Comparative Study of Nonlinear Machine Learning for the “In Silico” Depiction of Tyrosinase Inhibitory Activity from Molecular Structure

A Comparative Study of Nonlinear Machine Learning for the “In Silico” Depiction of Tyrosinase Inhibitory Activity from Molecular Structure

Atom based linear index descriptors in QSAR-machine learning classifiers for the prediction of ubiquitin-proteasome pathway activity

Exploring different strategies for imbalanced ADME data problem: case study on Caco-2 permeability modeling

Contact Info

Product

Resources

About