Artificial intelligence in human immunodeficiency virus mutation prediction and drug design: Advancing personalized treatment and prevention

Abstract

Despite significant advancements in highly active antiretroviral therapy (HAART), Human Immunodeficiency Virus (HIV) remains a global health challenge due to its high mutation rate, drug resistance, and the complexity of treatment optimization. Artificial intelligence (AI) has emerged as a transformative tool in HIV research, offering innovative solutions for predicting viral mutations, optimizing drug discovery and formulation design. However, challenges such as limited access to diverse datasets, ethical concerns, and model interpretability hinder the full potential of AI in HIV research. This review highlights gaps in AI-driven HIV research and explores advancements to address these challenges. AI-driven platforms, such as DeepHIV and geno2pheno, have demonstrated success in forecasting resistance mutations and guiding therapeutic decisions. AI is also revolutionizing drug formulation development by enhancing solubility, bioavailability, and stability, while improving patient adherence through advanced delivery systems. Current applications of AI in HIV mutation prediction, drug discovery, and formulation optimization have highlighted the potential of AI towards HIV management and eradication while addressing gaps in data availability and model transparency. By integrating structural, pharmacological, and clinical data, AI provides a comprehensive framework for rational drug design and personalized treatment strategies. By leveraging AI-driven insights, HIV treatment and prevention can become more personalized, efficient, and sustainable. Future research should focus on overcoming data limitations, enhancing model interpretability, and exploring innovative AI approaches to contribute to the global fight against the HIV epidemic.