Abstract
The COVID-19 pandemic highlighted critical limitations in the speed, scalability and translational efficiency of conventional antiviral drug discovery. Although vaccines and repurposed antivirals have reduced disease severity, the continued emergence of SARS-CoV-2 variants and breakthrough infections underscores the need for sustained discovery of novel therapeutics. The main protease (Mpro), an essential and highly conserved enzyme required for viral replication remains a validated and attractive antiviral target. Medicinal plants represent a vast and underexplored source of structurally diverse bioactive compounds with antiviral potential; however, traditional plant-based drug discovery approaches are often constrained by reliance on ethnobotanical knowledge and fragmented screening workflows. This review critically examines emerging strategies that integrate machine learning, LC–MS-based metabolomics and network pharmacology to modernise medicinal plant-based antiviral discovery. We highlight how machine learning enables data-driven prioritisation of candidate compounds and plant species beyond well-studied taxa, while metabolomics provides experimental validation through comprehensive chemical profiling and dereplication. Molecular docking and molecular dynamics further refine candidate selection by evaluating binding modes and stability, whereas network pharmacology offers systems-level insight into multitarget and multipathway effects. Importantly, we discuss key limitations of these approaches, including data bias, model interpretability, and gaps between in silico prediction and experimental validation. By synthesising these methodologies into a unified computational-experimental pipeline, this review provides a critical framework for accelerating the discovery of plant-derived Mpro inhibitors and supports the development of resilient antiviral strategies for current and future pandemics.