Differential evolution-binary particle swarm optimization algorithm for the analysis of aryl β-diketo acids for HIV-1 integrase inhibition

A hybrid differential evolution-binary particle swarm optimization (DE-BPSO) algorithm is proposed as a feature selection algorithm in the development of quantitative structure-activity relationship (QSAR) models. DE is used to evolve the velocities of the particle swarm from which a series of rules...

Full description

Saved in:
Bibliographic Details
Main Authors: Ko, G. M., Srinivas Reddy, A., Kumar, S., Garg, R., Bailey, B. A., Hadaegh, A. R.
Format: Conference Proceeding
Language:English
Subjects:
Analytical models
Biological system modeling
Chemicals
Compounds
Computational modeling
Optimization
Vectors
Online Access:Request full text
Tags: Add Tag
No Tags, Be the first to tag this record!
Description
Summary:A hybrid differential evolution-binary particle swarm optimization (DE-BPSO) algorithm is proposed as a feature selection algorithm in the development of quantitative structure-activity relationship (QSAR) models. DE is used to evolve the velocities of the particle swarm from which a series of rules are used to determine the discrete values of the position vectors which form chemical descriptor subsets. These descriptor subsets are then used to develop models for QSAR analysis. DE-BPSO was found to outperform the standalone BPSO algorithm. The DE-BPSO algorithm was then used to develop multiple linear regression models for the analysis of aryl β-diketo acid compounds for the inhibition of HIV-1 integrase. This model highlights the significance of hydrophobicity and partial positive charges of the hydrogen atoms on the molecular surface in influencing the biological activities of these compounds for the inhibition of HIV-1 integrase.
ISSN:1089-778X
1941-0026
DOI:10.1109/CEC.2012.6256578