🧬 BioBenchmarks
LIVE v2.0.0
Catalog, rank, and compare AI/ML benchmarks across the drug-discovery pipeline, plus the experts, groups, and initiatives behind them.
Virtual Cell → Clinical Development. Dual-purpose: what to benchmark against and who to collaborate with. Honest, differentiated 7-criterion rubric plus experimental validation tier. Last refreshed 2026-05-12.
90
Individual benchmarks
31
Initiatives
1,749
Benchmarks tracked across initiatives
2,413
Direct-linked hosted benchmarks
87
Experts
79
Labs / companies / consortia
28
Known-private benchmarks
12
Pipeline stages covered
What's new in v2.0
- Every benchmark now has an experimental validation tier — clinical / wet-lab-confirmed / prospective / retrospective
- 2,413 direct-linked hosted benchmarks across initiatives (Insilico portals, TDC, Polaris, ProteinGym, HuggingFace, MoleculeNet, DeepChem, DREAM, CASP)
- New Private benchmarks catalog — 28 industry benchmarks with gated access
- Benchmark comparison (up to 4 side-by-side), CSV export of any filtered view, keyboard shortcuts
- +17 newly catalogued benchmarks (v1: 73 → v2: 90)
Experimental validation distribution
Clinical (7)
Wet-lab confirmed (20)
Prospective (7)
Retrospective (56)
Top benchmarks (overall)
7-criterion weighted composite. Full rankings on /benchmarks.
| # | Benchmark | Stages | Validation | Score | Flags |
|---|---|---|---|---|---|
| 1 | Open Targets Platform | Disease ModelingTarget ID | Wet-lab confirmed | 100.0 | |
| 2 | DepMap (Cancer Dependency Map) | Target IDDisease Modeling | Wet-lab confirmed | 100.0 | |
| 3 | TDC ADMET Group | Lead ID / ADMET | Retrospective | 100.0 | |
| 4 | SAbDab | Hit IDLead ID / ADMETDevelopmental Candidate | Retrospective | 100.0 | |
| 5 | Protein Language Model Eval 2026 | Virtual CellHit ID | Wet-lab confirmed | 100.0 | |
| 6 | ChEMBL | Hit IDLead ID / ADMET | Wet-lab confirmed | 97.5 |
🧬 Aging / Longevity spotlight
Benchmarks relevant to geroscience and longevity therapeutics.
| # | Benchmark | Stages | Validation | Score |
|---|---|---|---|---|
| 1 | Longevity Benchmark (Insilico) | Disease ModelingTarget IDPost-market / RWE | Prospective | 90.6 |
| 2 | Longevity Compound Benchmark | Hit IDLead ID / ADMET | Wet-lab confirmed | 84.6 |
| 3 | MoleculeNet | Lead ID / ADMETHit ID | Retrospective | 78.0 |
| 4 | SIDER | Post-market / RWEIND-enabling | Retrospective | 74.9 |
Top initiatives (meta-platforms)
| # | Initiative | Tracked | Direct links | Score |
|---|---|---|---|---|
| 1 | HuggingFace — Bio/Chem Datasets | 310 | 341 | 67.8 |
| 2 | ScienceAIBench | 227 | 227 | 90.6 |
| 3 | ProteinGym | 217 | 283 | 97.5 |
| 4 | Drug Discovery Benchmarks (DDB) | 206 | 206 | 87.6 |
| 5 | InsilicoBench | 162 | 162 | 84.6 |
| 6 | Papers With Code — Drug Discovery | 120 | 0 | 71.6 |
Top experts & groups
Experts
| # | Expert | Affiliation | Score |
|---|---|---|---|
| 1 | Marinka Zitnik | Harvard Medical School (Zitnik Lab) | 100.0 |
| 2 | Charlotte Deane | Oxford (OPIG) / UKRI Executive Chair | 100.0 |
| 3 | Torsten Schwede | Biozentrum Basel / SIB | 100.0 |
| 4 | Alex Zhavoronkov | Insilico Medicine | 100.0 |
| 5 | Fabian Theis | Helmholtz Munich / TUM | 96.2 |
Groups / Labs / Companies
| # | Group | Kind | Score |
|---|---|---|---|
| 1 | Broad Institute | lab | 100.0 |
| 2 | Oxford OPIG (Deane Lab) | lab | 100.0 |
| 3 | EMBL-EBI | consortium | 97.4 |
| 4 | MIT CSAIL / Jameel Clinic / Coley Lab | lab | 97.4 |
| 5 | Zitnik Lab | lab | 96.9 |
Keyboard shortcuts ·
/ focus search ·
g then: h home · b benchmarks · s stages · v validation · e experts · g groups · i initiatives · m matrix · p private · d downloads · a about