Benchmark RadarShallowBench
ShallowBench is a curated benchmark of 5,780 shallow-pocket targets for evaluating generative drug design models. Targets are extracted from CrossDocked2020 based on low concavity and sufficient surface area.
- Released
- 2026-06-04
- Readiness
- Paper only
- Primary field
- Health & Life Sciences
Why it matters
Generative models often rely on deep pockets and struggle with shallow pockets. ShallowBench provides a testbed for developing models that can handle challenging targets.
Motivation
While generative AI models have demonstrated remarkable success in structure-based drug design, they predominantly rely on deep binding pockets and struggle to sample effective ligands for challenging low-pocketability targets, such as the historically "undruggable" oncology targets KRAS and MYC.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.