| name | download-ncbi-data |
| description | Use when downloading static NCBI reference datasets such as taxonomy, MeSH tree, bioconcepts, generif, journals, serials, or PMC open access files via CLI. |
| disable-model-invocation | true |
| user-invocable | true |
download-ncbi-data
Quick Start
- Command:
download-ncbi-data <dataset> [extra-arg]
- Local executable:
/home/vimalinx/miniforge3/envs/bio/bin/download-ncbi-data
- Full reference: See
references/help.md
When To Use This Tool
- Download curated NCBI side datasets such as taxonomy tables, GeneRIF summaries, journal lists, MeSH resources, PMC OA bundles, or NIH Open Citation Collection files.
- Build local lookup tables without manually browsing NCBI FTP trees.
- Grab sample/demo payloads like
carotene, globin, human, or smear for testing downstream EDirect tooling.
Common Patterns
download-ncbi-data taxnames
download-ncbi-data meshtree
download-ncbi-data journals
download-ncbi-data nihocc
download-ncbi-data oa-book NBKXXXXXX
Recommended Workflow
- Choose a dedicated output directory before running the command because many subcommands emit multiple derived files.
- Pick the dataset subcommand deliberately:
taxnames/taxoninfo for taxonomy, meshtree for MeSH, serials/journals for literature metadata, pmc-oa or pmc-bioc for large content bundles.
- Let the script finish its side products, such as
taxnames.txt, lineages.txt, meshconv.xml, meshtree.txt, or journal lookup tables.
- Treat the generated files as reusable local resources for later pipelines instead of re-downloading them each run.
Guardrails
- This tool writes into the current directory and may create several companion files in addition to the primary download.
oa-book expects an accession argument after the subcommand; most other modes do not.
- Some options are real datasets, while
carotene, globin, human, and smear are sample payloads meant for testing.
- The local script has no meaningful
--version mode and depends heavily on live network access.
- In this local install, the
human sample branch appears to contain a filename typo in the wrapper script, so verify that mode manually before depending on it.