BioMysteryBench
An agentic bioinformatics benchmark of objective, expert-authored mysteries over anonymized real-world biological data, scored on final answers rather than prescribed analysis paths.
Scientific domain
Single-cell molecular measurements and analyses.
单细胞组学
An agentic bioinformatics benchmark of objective, expert-authored mysteries over anonymized real-world biological data, scored on final answers rather than prescribed analysis paths.
A containerized benchmark of long-horizon bioinformatics analysis over real published notebooks and associated data, with open-answer and multiple-choice evaluation modes.
A 100-task agent benchmark of objectively gradable computational-biology problems requiring multi-step reasoning, bespoke code, tools, and real-world external resources.
A reusable benchmark for comparing differential transcript usage detection tools across simulated and real transcriptomics data.
Real single-cell RNA-seq data augmented with known gene perturbations for comparing feature-selection methods.
A research-level agent benchmark of 129 synthetic, multistage computational-biology analyses that require iterative QC, statistical modeling, diagnostics, and decision-relevant judgment.
Agentic evaluation suite for data-grounded single-cell analysis across diverse sequencing technologies and workflow stages.
A 13-task benchmark of single-cell data integration across simulated, scRNA-seq, and scATAC-seq settings, evaluated with 14 batch-removal and biological-conservation metrics.
A benchmark for evaluating LLM cell-type annotation across scRNA-seq and single-cell multiomics data.
A benchmark of deterministic, verifiable agentic problems derived from real spatial-transcriptomics workflows, testing whether agents can manipulate data and recover key biological results.
Registry records tagged Single-cell, counted by capability.
| Capability | Records |
|---|---|
| Retrieval | 1 |
| Classification | 1 |
| Data analysis | 9 |
| Coding | 6 |
| Tool use | 6 |
| Scientific reasoning | 7 |
Task mappings are evidence-backed and may be partial for mixed suites.