genomebundle

genomebundle bundles metadata and files for Earth BioGenome Project (EBP) genome assemblies into a single, reproducible package.

The core idea is simple: when you use a genome assembly in your research, you should be able to document exactly what you downloaded, when, and from where — with checksums. genomebundle does this by aggregating data from three sources into a machine-readable manifest.json:

• GoaT (Genomes on a Tree) — taxonomy and cross-references
• NCBI Datasets — assembly statistics and FTP file URLs
• BlobToolKit — BUSCO completeness results (assembly quality metrics)

This makes it easier to cite the data precisely and to keep pipelines reproducible across time.

Installation

pip install genomebundle

Basic CLI usage

# Download FASTA and GFF
genomebundle fetch GCF_040938575.1 --files fasta,gff

# Download all associated files
genomebundle fetch GCF_040938575.1 --files all

# Build manifest only (no download)
genomebundle fetch GCF_040938575.1 --no-download

# Verify checksums of a downloaded bundle
genomebundle verify ./GCF_040938575.1/

# Print manifest of an existing bundle
genomebundle show ./GCF_040938575.1/

Python API

from genomebundle import fetch_assembly, fetch_assembly_report, fetch_busco

goat = fetch_assembly("GCF_040938575.1")
ncbi = fetch_assembly_report("GCF_040938575.1")
btk  = fetch_busco("GCF_040938575.1")

Output

Each bundle contains:

• manifest.json — machine-readable, includes SHA256 checksums and source URLs
• README.txt — human-readable summary
• downloaded files (optional)

References

• Challis et al. (2023). GoaT: Genomes on a Tree. Wellcome Open Research. https://doi.org/10.12688/wellcomeopenres.18658.1
• Byrd et al. (2024). Best practices for genetic and genomic data archiving. Nature Ecology & Evolution. https://doi.org/10.1038/s41559-024-02423-7
• Dainat et al. (2025). Guidelines for gene and genome assembly nomenclature. Genetics. https://doi.org/10.1093/genetics/iyaf006

License

MIT