What problem does it solve?
This Skill helps manage and analyze biological datasets by providing a unified platform for data management, annotation, ontologies, schema validation, and integration with workflow managers and MLOps platforms.
Core Features & Use Cases
- Data Management: Manage biological datasets, track computational workflows, curate and validate data with biological ontologies.
- Schema Validation: Validate datasets against predefined schemas.
- Integration: Integrates with workflow managers (Nextflow, Snakemake, Redun) and MLOps platforms (W&B, MLflow, HuggingFace, scVI-tools).
- Use Case: For a single-cell RNA-seq analysis, LaminDB can be used to track data lineage, manage annotations, and integrate with ML pipelines for model training and validation.
Quick Start
To start using LaminDB, first install it and initialize an instance:
pip install lamindb
lamin init --storage ./mydata
Then, import your datasets and start tracking your workflows:
ln track
ln Artifact.get(key="path/to/your/data.h5ad").load()
# ... your analysis code ...
ln finish