Researcher
Background
A principal investigator running a computational biology lab with 8 researchers, 3 AI agents, and growing experimental data. Publishes 10+ papers annually. Generates protocols, datasets, and analysis pipelines faster than the lab can organize them. New postdocs spend their first month just figuring out where things are.
Goals
- Organize knowledge sprawl across multiple platforms into a single navigable structure
- Onboard new postdocs in days, not weeks
- Make AI agents lab-specific: load domain context so output reflects actual lab methods, not generic biology
- Enable protocol reusability and analysis pipeline sharing with collaborating labs
Pain Points
- Knowledge scattered across Slack, Google Docs, Jupyter notebooks, and shared drives
- File naming chaos (“data_final_v3_REAL.csv”) with no conventions
- New postdocs spend their first month finding things rather than doing science
- AI agents produce generic output without understanding lab-specific methods and protocols
How They Use aDNA
Heavy ontology extension to capture domain-specific knowledge types:
what/context/— 8 domain subtopics (protein folding, binding affinity, molecular docking, etc.), each 150-300 lines, quality-scored with token estimateswhat/protocols/— 25 experimental protocols as structured documents (ontology extension)what/datasets/— dataset metadata with lineage tracking and FAIR metadata (ontology extension)what/lattices/— analysis pipelines as composable YAML lattice definitionshow/campaigns/— multi-month research projects decomposed into experiment-level missionshow/sessions/— audit trail linking agent work to specific experiments- Federation — sharing validated analysis lattices with collaborating labs
Self-reference: This vault’s context library (what/context/ with 5 topics, 27 subtopics, ~75K tokens) mirrors what the research lab builds for their domain. The structure — topic directories with AGENTS.md indexes and token estimates — is identical; only the subject matter differs.
Typical Ontology Extensions
| Entity | Triad | Purpose |
|---|---|---|
protocol | what/ | Experimental protocols (wet lab + computational) |
dataset | what/ | Dataset metadata, lineage, and FAIR annotations |
experiment | how/ | Individual experiment records with methods and results |
collaboration | who/ | Cross-lab coordination and data sharing agreements |
Related
- Research Lab Use Case — full narrative
- Tutorial: Extend the Ontology — adding domain entity types
- FAIR Metadata (concept) — the metadata standard for research data
- Federation Readiness Pattern — preparing for cross-lab sharing