Search

Background: Coxsackievirus group B (CVBs) are common enteroviruses associated with several diseases from etiologically to inflammatory cardiomyopathies and constitute a severe cause of mortality in newborn resulting in severe meningitis, fulminant infection, myocarditis, and encephalitis. While Berberian (BBR) is an effective antivirus and possesses potentials of suppressing CVB replication, Zeng et al. explored a structural modification of BBR by incorporating a substituted primary amine enhance antiviral potency and safety. Based on data set from Zeng et al., we attempted to propose a QSAR model that can predict the bioactivity of unknown compounds as anti-CVB1. Results: Among many descriptors, four were selected using the Genetic Functional Approximation (GFA). Internal and external validation was carried out on data set using statistical parameters. The QSAR model was seen to meet the minimum requirement with Lack of fit = 0.068744, R2 0.897, Adjusted R2 = 0.8627, cross-validated R2 = 0.76169, R2 predicted = 0.68. Conclusion: The predictive ability of the model was found to be satisfactory and could be used for designing a similar group of compounds. [ABSTRACT FROM AUTHOR]

Background: Coxsackievirus group B (CVBs) are common enteroviruses associated with several diseases from etiologically to inflammatory cardiomyopathies and constitute a severe cause of mortality in newborn resulting in severe meningitis, fulminant infection, myocarditis, and encephalitis. While Berberian (BBR) is an effective antivirus and possesses potentials of suppressing CVB replication, Zeng et al. explored a structural modification of BBR by incorporating a substituted primary amine enhance antiviral potency and safety. Based on data set from Zeng et al., we attempted to propose a QSAR model that can predict the bioactivity of unknown compounds as anti-CVB1. Results: Among many descriptors, four were selected using the Genetic Functional Approximation (GFA). Internal and external validation was carried out on data set using statistical parameters. The QSAR model was seen to meet the minimum requirement with Lack of fit = 0.068744, R2 0.897, Adjusted R2 = 0.8627, cross-validated R2 = 0.76169, R2 predicted = 0.68. Conclusion: The predictive ability of the model was found to be satisfactory and could be used for designing a similar group of compounds. [ABSTRACT FROM AUTHOR]