Login / Signup

A binary QSAR model for classifying neuraminidase inhibitors of influenza A viruses (H1N1) using the combined minimum redundancy maximum relevancy criterion with the sparse support vector machine.

M K QasimZakariya Y AlgamalH T Mohammad Ali
Published in: SAR and QSAR in environmental research (2018)
Quantitative structure-activity relationship (QSAR) classification modelling with descriptor selection has become increasingly important because of the existence of large datasets in terms of either the number of compounds or the number of descriptors. Descriptor selection can improve the accuracy of QSAR classification studies and reduce their computation complexity by removing the irrelevant and redundant descriptors. In this paper, a two-stage classification approach is proposed by combining the minimum redundancy maximum relevancy criterion with the sparse support vector machine. The experimental results of classifying the neuraminidase inhibitors of influenza A (H1N1) viruses show that the proposed method is able to effectively outperform other sparse alternatives methods in terms of classification performance and the number of selected descriptors.
Keyphrases
  • deep learning
  • structure activity relationship
  • machine learning
  • molecular docking
  • molecular dynamics
  • high resolution
  • neural network
  • genetic diversity