Stage: Target ID

19 benchmarks, ranked by composite score. ← All stages

#BenchmarkDescriptionModalitiesScoreFlags
1 Open Targets Platform Integrates genetic, genomic, pharmacological evidence to score targets for 20k+ diseases. cross-modality 100.0
2 DepMap (Cancer Dependency Map) Genome-scale CRISPR/RNAi essentiality across 1150 cancer cell lines + omics. cross-modality 100.0
3 ProteinGym 217 DMS substitution assays + indel + clinical variants — de facto standard for VEPs. protein-general 97.5
4 STRING Protein-protein interaction & functional association network across 12k organisms. protein-general 94.9
5 CASP15 CASP15 blind structure prediction including multimer and ligand-bound categories. protein-general 94.9
6 PrimeKG Precision medicine knowledge graph integrating 20 sources for 17k diseases. cross-modality 91.9
7 Longevity Benchmark (Insilico) 19 aging/longevity benchmarks on ScienceAIBench/InsilicoBench/DDB — NHANES mortality, TCGA survival, methylation age, GTEx, Olink proteomic, longevity synergy. cross-modality 90.6
8 LINCS L1000 / CMap 1.3M transcriptional profiles across 978 landmark genes after genetic or chemical perturbations. small-moleculecross-modality 89.9
9 canSAR Cancer translational knowledge base integrating pharmacology, bioactivity, structure, druggability. small-moleculeprotein-general 89.4
10 scPerturb Harmonized single-cell perturbation datasets (genetic + chemical) with standardized metadata. cross-modality 88.9
11 CZ Virtual Cell Challenge Open challenge to predict transcriptomic responses to genetic & chemical perturbations across cell types, hosted by CZ Biohub. cross-modality 88.1
12 TargetBench (Insilico) 10 disease-area target identification benchmarks on ScienceAIBench/DDB (cancer, cardiovascular, endocrine/metabolic, fibrotic, inflammation/immunology, neuro, etc.). cross-modality 84.6
13 CAFA5 CAFA 5th edition — blind GO annotation eval (142k targets; Kaggle 2023, 1625 teams). protein-general 84.3
14 DisGeNET Large collection of gene-disease associations (text-mining + curation + repositories). cross-modality 81.0 license-gated-commercial
15 FLIP Fitness landscape inference benchmarks with realistic train/test splits (AAV, GB1, Meltome, SCL, Bind). protein-general 80.8
16 CPTAC Proteogenomic Benchmarks Proteogenomic benchmarks across 10 tumor types — drives DREAM proteogenomic challenges. cross-modality 80.8
17 TAPE Tasks Assessing Protein Embeddings — 5 tasks (secondary structure, contact, fluorescence, stability, homology). protein-general 74.9 deprecated-recommend-replace
18 PEER PEER — 14 protein property prediction tasks across structure/function/interaction. protein-general 74.4
19 ClawBio Skill Correctness Bench Third-party (Biostochastics LLC) benchmark of bio-analysis skills on safety / correctness / honesty. 10 skills × 182 tests. cross-modality 74.2