Development of classification models for identification of important structural features of isoform-selective histone deacetylase inhibitors (class I)

Neiband, Marzieh Sadat; Benvidi, Ali; Mani-Varnosfaderani, Ahmad

doi:10.1007/s11030-019-10013-0

Cited by 5 publications

(12 citation statements)

References 58 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…To prepare the molecules for analysis, several steps were taken. These steps included adding hydrogen atoms to the molecules, removing similar structures, generating 3D structures, and performing 3D optimization of molecular structures using the Merck Molecular Force Field (MMFF94) to determine partial atomic charges 20 . These data preparation steps were conducted using Open Babel software, version 2.4.0 22 .…”

Section: Methodsmentioning

confidence: 99%

“…S1 c outlines the steps in model development procedure. After scaling the dataset it was randomly partitioned into 70% for training and cross-validation, and 30% for the test set 20 . In this study, to ensure that the test set data were not highly similar to the training set data, a similarity analysis was conducted between test set and training set molecules.…”

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%

“…The data set was randomly divided into 70% calibration (training) and 30% test set to validate the models 20 , as detailed in section “ Model development ”. The models were established using the training set compounds and were subsequently verified for prediction accuracy using the independent test set 20 . The stability and accuracy of the classification models were assessed using tenfold Venetian-Blinds cross-validation on the training set molecules 20 .…”

Section: Methodsmentioning

confidence: 99%

“…To assess the molecular characteristics of each ligand, a comprehensive range of molecular descriptors were computed 18 . Furthermore, we employed two machine learning methods, namely the counter propagation artificial neural network (CPANN) 19 and supervised Kohonen networks (SKN) 20 for classification of ligands based on their activities and therapeutic targets. The findings of this study demonstrate a clear correlation between the selected descriptors and the activity/selectivity of molecules, providing valuable insight that can aid medicinal chemists in developing drugs with improved efficacy and reduced side effects.…”

Section: Introductionmentioning

confidence: 99%

See 4 more Smart Citations

Deriving general structure–activity/selectivity relationship patterns for different subfamilies of cyclin-dependent kinase inhibitors using machine learning methods

Kaveh,

Mani-Varnosfaderani,

Neiband

2024

Sci Rep

View full text Add to dashboard Cite

Cyclin-dependent kinases (CDKs) play essential roles in regulating the cell cycle and are among the most critical targets for cancer therapy and drug discovery. The primary objective of this research is to derive general structure–activity relationship (SAR) patterns for modeling the selectivity and activity levels of CDK inhibitors using machine learning methods. To accomplish this, 8592 small molecules with different binding affinities to CDK1, CDK2, CDK4, CDK5, and CDK9 were collected from Binding DB, and a diverse set of descriptors was calculated for each molecule. The supervised Kohonen networks (SKN) and counter propagation artificial neural networks (CPANN) models were trained to predict the activity levels and therapeutic targets of the molecules. The validity of models was confirmed through tenfold cross-validation and external test sets. Using selected sets of molecular descriptors (e.g. hydrophilicity and total polar surface area) we derived activity and selectivity maps to elucidate local regions in chemical space for active and selective CDK inhibitors. The SKN models exhibited prediction accuracies ranging from 0.75 to 0.94 for the external test sets. The developed multivariate classifiers were used for ligand-based virtual screening of 2 million random molecules of the PubChem database, yielding areas under the receiver operating characteristic curves ranging from 0.72 to 1.00 for the SKN model. Considering the persistent challenge of achieving CDK selectivity, this research significantly contributes to addressing the issue and underscores the paramount importance of developing drugs with minimized side effects.

show abstract

Section: Methodsmentioning

confidence: 99%

Section: Methodsmentioning

confidence: 99%