Computational approaches for flavivirus drug discovery: targeting structural and non-structural proteins

The Flaviviridae family comprises some of the most pathogenic viruses associated with global outbreaks and severe health complications. Mosquito-borne viruses such as Zika, Dengue, West Nile, Japanese Encephalitis, and Yellow Fever account for millions of infections annually, particularly in tropical and subtropical regions. Their ability to infect multiple tissues and cause a wide range of clinical manifestations poses a significant global public health challenge. Despite progress in developing therapeutic strategies and vaccines, effective antiviral treatments remain limited, which emphasises the urgent need to identify novel lead compounds that act through distinct inhibitory mechanisms. Understanding the structural and functional organisation of flavivirus proteins, including the capsid, pre-membrane, envelope, and seven non-structural proteins, can facilitate the identification of promising therapeutic targets. However, frequent mutations in these proteins often confer resistance, underscoring the need for more robust antiviral strategies. Computational approaches offer powerful tools for accelerating antiviral drug discovery by enabling accurate prediction of biological activity and toxicity from chemical structures. Integrating these methods into drug development pipelines enables the rapid identification and optimisation of promising candidates. Accordingly, this review discusses recent applications of computational techniques in the discovery of flavivirus inhibitors, highlighting key examples and emerging trends from recent studies. It also outlines current challenges and future directions in advancing flavivirus drug discovery.