What problem does it solve?
LaminDB helps you manage biological datasets so they become queryable, traceable, reproducible, and FAIR by capturing lineage from inputs to outputs and enforcing structured metadata through typed features and schema validation.
Core Features & Use Cases
- Data lineage tracking: Record computational workflows (runs and transforms) so results can be traced back to inputs, code, parameters, and environments.
- Queryable artifacts with typed metadata: Store datasets as versioned artifacts and annotate them with typed features for efficient filtering and discovery.
- Schema validation & ontology-based curation: Validate datasets against flexible/minimal/strict schemas and standardize annotations using biological ontologies (e.g., via bionty).
- Common use cases: scRNA-seq and spatial transcriptomics curation, data lakehouse-style querying across experiments, and integration with workflow managers (Nextflow/Snakemake) and MLOps tools (W&B/MLflow).
Quick Start
Use LaminDB to validate and ontology-standardize your dataset by curating your DataFrame with a schema and saving it as a linked, versioned artifact.