What problem does it solve?
LaminDB helps research teams turn biological datasets and analysis pipelines into queryable, traceable, and reproducible assets with FAIR-aligned metadata and lineage tracking.
Core Features & Use Cases
- Data lineage & provenance: Track which inputs and code produced which outputs across notebook/script workflows.
- FAIR curation with typed features: Validate and annotate datasets using schema-defined, ontology-linked metadata for reliable querying.
- Ontology-driven biological annotation: Standardize genes, cell types, tissues, diseases, and other entities via controlled vocabularies (e.g., bionty/Bionty) to reduce inconsistency.
- Scalable data lakehouse patterns: Organize artifacts across storage backends and query them by metadata without repeatedly reloading raw files.
- Workflow & MLOps integration: Connect curated artifacts to workflow managers (Nextflow/Snakemake/Redun) and experiment tracking platforms (W&B/MLflow) to keep experiments, code, and data in sync.
Quick Start
Use the lamindb skill to curate and validate an scRNA-seq AnnData file by defining schemas and linking ontology terms, then save the validated artifact with preserved provenance.