1st International and 10th National Iranian Conference on Bioinformatics
Home Accepted Papers
Identification of molecular features and pharmacophores for selective inhibition of cyclin-dependent kinases: Application of counterpropagation artificial neural networks
Paper ID : 1479-ICB10
Authors:
Sara Kaveh1, Marzieh Sadat Neiband2, Ahmad Mani-Varnosfaderani *1
1Department of Chemistry, Tarbiat Modares University, Tehran, Iran
2Department of Chemistry, Payame Noor University (PNU), Tehran, Iran
Abstract:
Protein kinases are one of the largest enzyme families, consisting of 2% of the translated human genome. Kinase inhibitors have enormous medical use for treating mortal diseases such as cancer and Parkinson. Due to the wide range of activities and functions for all kinases, there is a vital need for the development of selective kinase inhibitors for targeting only some specific types of them for effective treatments [1]. The main aim of this project is to find isoform-selective pharmacophores and molecular features for different types of Kinase receptors. In order to achieve this goal, a total of 4201 drug-like molecules with recorded inhibition activity for the inhibition of CDK1, CDK2, CDK4, CDK5, and CDK9, were collected from Binding-Database [2]. The variable importance in projection (VIP) method was used to select the molecular descriptors from the 3224 descriptor pool calculated via DRAGON 5.5 software [30]. The dataset was divided into training (70%) and test (30%) sets, randomly. Counter propagation artificial neural network (CPANN) and supervised Kohonen networks (SKN), were used for the classification of the molecules. Some general parameters such as mean square distance index, number of Pyrazoles atoms, hydrophobicity, aromaticity index, and number of hydroxyl groups were found to be important parameters for describing the inhibition behavior of CDK’s inhibitors. Generally, the performances of classification models were evaluated according to the statistical parameters derived from the confusion matrices [3,4]. The classification rates range from 82 % to 79% for the training and validation procedure for the optimized CPANN models. The high accuracy values of the obtained classifiers for the training and test sets demonstrate that the information provided is reliable for describing and predicting the activity of CDK inhibitors. The reliable statistical values of the classified models can be applied by researchers in the pharmaceutical sciences whom aim to design selective kinase inhibitors [3,4].
Keywords:
Classification; Kinase; Selective drug design; structure-activity relationship; Virtual screening
Status : Paper Accepted (Poster Presentation)