Fragment-based drug discovery: A graphical review

Highlights

• FBDD identifies low-MW fragments by biophysical/structural methods and optimises them to leads.
• FBDD samples chemical space efficiently; >50 fragment-derived compounds entered clinical development.
• Computational tools (docking, FEP, AI/ML) guide fragment growth, complementing screens and speeding optimisation.
• Hybrid platforms combine biophysical and AI/ML to enhance hit discovery and filter artefacts.
• FBDD has delivered approved drugs (e.g., Vemurafenib, Venetoclax) and advanced clinical candidates.

Abstract

Three decades after its introduction, fragment-based drug (or lead) discovery (FBDD or FBLD) has become a mature and powerful strategy for generating novel leads, offering distinct advantages for challenging or previously “undruggable” targets where traditional screening (e.g., high throughput screening) often fails. The FBDD approach identifies low molecular weight fragments (MW < 300 Da) that bind weakly to a target; these interactions are detected using highly sensitive biophysical methods such as NMR, X-ray crystallography, and SPR. These initial hits are then optimised into potent leads through structure-guided strategies, including fragment growing, linking, or merging. This graphical review illustrates the modern FBDD workflow, highlighting the critical integration of experimental and computational methods. We discuss how innovations in library design, hybrid screening platforms, and the application of AI/ML are accelerating discovery cycles and improving hit validation. The power of this approach is demonstrated through case studies of FDA-approved drugs, including Vemurafenib and Venetoclax, which progressed from simple fragments to transformative medicines. Finally, we provide an outlook on the future of FBDD as it continues to evolve with emerging technologies to push the boundaries of drug discovery.