Targeting HIV/HCV Coinfection Using a Machine Learning-Based Multiple Quantitative Structure-Activity Relationships...

Area of Science:

• Computational chemistry and drug discovery
• Infectious diseases
• Pharmacology

Background:

• Human immunodeficiency virus type-1 (HIV-1) and hepatitis C virus (HCV) coinfection presents treatment challenges due to hepatic safety and drug-drug interaction concerns.
• Developing multitarget inhibitors is a promising strategy for HIV/HCV coinfection, but experimental identification is costly and time-consuming.

Purpose of the Study:

• To develop and validate in silico models for predicting multitarget inhibitors against HIV-1 and HCV.
• To identify novel compounds with potential activity against both HIV-1 and HCV targets using computational methods.

Main Methods:

• Construction of 60 classification models using Naïve Bayes (NB) and Support Vector Machine (SVM) algorithms with MACCS and ECFP6 molecular fingerprints.
• Application of a multiple quantitative structure-activity relationships (multiple QSAR) method to predict compound activity against 11 HIV-1 and 4 HCV targets.
• Validation of models using five-fold cross-validation and test set validation, followed by prediction on additional compounds.

Main Results:

• The multiple QSAR models demonstrated high classification accuracy, with Area Under the ROC Curve (AUC) values ranging from 0.83 to 1 (mean 0.97 for HIV-1, 0.96 for HCV).
• Prediction on 46 compounds identified 20 potential multitarget inhibitors, including approved HIV-1 and HCV drugs, and novel compounds.
• Experimental validation confirmed that 7 out of 9 tested compounds interacted with both HIV-1 and HCV targets.

Conclusions:

• The multiple QSAR method is effective for predicting chemical-protein interactions and discovering multitarget inhibitors.
• This computational approach offers a cost-effective and efficient strategy for developing novel therapeutics for HIV/HCV coinfection.
• Further experimental investigation of predicted hits is warranted to advance the treatment of HIV/HCV coinfection.