Benchmark RadarChem2Gen-Bench
Chem2Gen-Bench evaluates chemical-to-genetic translation using 260,084 chemical and 1,099,045 genetic perturbation profiles in cell-target contexts. It measures pairwise alignment, retrieval success, and representation quality across matched perturbation settings.
- Released
- 2026-06-19
- Readiness
- Paper only
- Primary field
- Health & Life Sciences
Why it matters
Chemical and genetic perturbations are often studied separately, leaving translation between them under-tested. This benchmark provides a comparative evaluation of retrieval and representation methods, offering decision value for selecting models that align perturbations around shared targets.
Motivation
Virtual-cell and perturbation models are increasingly used to predict cellular responses for biomedical discovery, but chemical and genetic perturbations are not automatically interchangeable.
Primary resources
Benchmark Radar records only publicly supported details and links back to primary sources for verification.