Correlation intensity index (CII) as a benchmark of predictive potential: Construction of quantitative structure activity relationship models for anti-influenza single-stranded DNA aptamers using Monte Carlo optimization

Abstract

The influenza virus threat continues to be a major global health issue. Although comprehensive studies concentrate on anti-influenza testing, however, the worldwide reports of fatalities caused by the influenza virus are still constant. The hemagglutinin viral protein (HA) is responsible for the first stage of the influenza virus, thus, it is considered a novel target. Single-stranded DNA aptamer could inhibit influenza infection via blocking of the receptor-binding region of the surface glycoprotein HA. So in the present study, we have developed robust QSAR models for 98 anti-influenza aptamers. Here, we have employed a statistical parameter, correlation intensity index (CII) to calculate the aptamers sequence-based optimal descriptors. The inbuilt Monte Carlo algorithm of CORAL software is used to build up the QSAR model using three target functions i.e. TF1 (IICweight=0.0 and CIIweight=0.0), TF2 (IICweight=0.3 and CIIweight=0.0) and TF3 (IICweight=0.0 and CIIweight=0.3). Models developed by considering correlation intensity index were found to be statistically more significant and robust. The developed QSAR model with TF3 having Rvalidation2=0.8801 for split 1 is considered as leading model. The common promoters of increase and decrease of endpoint were also extracted from three splits 1, 3 and 4.