bdsc-cli

Small CLI for syncing public Bloomington Drosophila Stock Center datasets and querying them locally.

Repo:

• https://github.com/gumadeiras/bdsc-cli

Primary source:

• https://bdsc.indiana.edu/stocks/stockdata.html

What it does:

• syncs BDSC CSV datasets into a local cache
• builds a local SQLite index
• supports local text search and stock lookups
• exposes optional live search against BDSC's current web endpoint

No third-party Python dependencies.

Install

Another computer:

brew tap gumadeiras/tap
brew install bdsc-cli

Or install the release wheel directly with pipx:

pipx install 'bdsc-cli @ https://github.com/gumadeiras/bdsc-cli/releases/download/v0.2.0/bdsc_cli-0.2.0-py3-none-any.whl'

Or with plain pip:

python3 -m pip install 'bdsc-cli @ https://github.com/gumadeiras/bdsc-cli/releases/download/v0.2.0/bdsc_cli-0.2.0-py3-none-any.whl'

Source install:

git clone https://github.com/gumadeiras/bdsc-cli.git
cd bdsc-cli
python3 -m pip install .

Repo-local dev install:

cd ~/git/bdsc-cli
python3 -m venv .venv
. .venv/bin/activate
python -m pip install -e .

Check the CLI:

bdsc --help

Build release artifacts locally:

python -m pip install -e .[release]
python -m build
python -m twine check dist/*
python scripts/render_homebrew_formula.py dist/bdsc_cli-$(python - <<'PY'
from bdsc_cli import __version__
print(__version__)
PY
).tar.gz

PyPI note:

• the GitHub release is live
• PyPI trusted publishing is not configured yet for bdsc-cli
• pip install bdsc-cli will work after that publisher is added

Quickstart

Create a local cache and index:

bdsc sync

Then query it:

bdsc find Chronos
bdsc find 'Or56a Lexa'
bdsc report optogenetics
bdsc find --gene Or56a --property lexA
bdsc find --gene Or42b --driver-family lexA
bdsc find RRID:BDSC_77118
bdsc find FBti0195688
bdsc stock 77118

Usage

Default state directory:

~/.local/share/bdsc-cli

Sync datasets and build the local index:

bdsc sync

Use find for nearly all interactive querying:

bdsc find Chronos
bdsc find Chronis
bdsc find 'Or56a Lexa'
bdsc find FBgn0003996 --json
bdsc find RRID:BDSC_77118
bdsc find FBti0195688
bdsc find --kind property VALIUM20
bdsc find --kind property-exact lexA
bdsc find --kind driver-family QF
bdsc find --kind relationship RNAi
bdsc find --gene Or56a --property lexA
bdsc find --gene Or42b --driver-family lexA
bdsc find --gene Or42b --driver-family qf
bdsc find --dataset genes --property olfactory --relationship coding --jsonl

Use canned reports for common retrieval buckets:

bdsc report olfactory
bdsc report drivers --jsonl
bdsc report optogenetics --limit 50

Inspect cache/index status:

bdsc status

Use a custom cache/index location:

bdsc sync --state-dir ./data
bdsc find Chronos --state-dir ./data

Structured output for scripts or agents:

bdsc status --json
bdsc find Chronos --json
bdsc find FBgn0003996 --dataset genes --json
bdsc find --gene Or56a --property lexA --json
bdsc export components --limit 5 --format jsonl
bdsc stock 77118 --json

Commands

• bdsc sync: download the BDSC CSV datasets; builds the index by default
• bdsc build-index: rebuild the SQLite index from previously downloaded CSVs
• bdsc status: show local dataset freshness and index metadata
• bdsc find [query]: primary query command; free-text lookup or compound filters
• bdsc report <name>: canned reports for olfactory, drivers, optogenetics
• bdsc export <dataset>: stream normalized rows as jsonl, csv, or tsv
• bdsc terms <scope>: inspect available property/relationship vocab
• bdsc stock <stknum>: local stock details
• legacy compatibility shims still exist for search, gene, component, fbid, rrid, property, property-exact, driver-family, relationship, lookup, filter, live-search

Find

Use find when the caller does not want to choose a dedicated query command up front.

Auto-detect rules:

• digits -> stock
• RRID:BDSC_* or BDSC_* -> rrid
• FBgn... -> gene
• FBti... / FBal... / similar FB.. ids in the component table -> fbid
• transgene/component-like text (P{...}, brackets, attP, CyO) -> component
• --kind property when you want property-driven lookup explicitly
• everything else -> gene, then local full-text search fallback if no gene hits

Examples:

bdsc find Chronos
bdsc find RRID:BDSC_77118
bdsc find --kind component 'P{10XUAS-Chronos'
bdsc find --kind property VALIUM20
bdsc find --kind property-exact lexA
bdsc find --kind driver-family qf

Export

Use export when another tool wants direct normalized rows instead of search-oriented output.

Datasets:

• stocks
• components
• genes
• properties

Examples:

bdsc export stocks --limit 3
bdsc export genes --query Chronos --kind gene
bdsc export components --query FBti0195688 --kind fbid --format jsonl
bdsc export properties --query VALIUM20 --kind property --format tsv
bdsc export components --gene Or56a --property lexA --format jsonl
bdsc export components --gene Or42b --driver-family qf --format jsonl
bdsc export genes --property olfactory --relationship coding --format csv
bdsc export components --format tsv --output components.tsv
bdsc export genes --format csv --output genes.csv
bdsc export properties --limit 20 --format jsonl

export --query uses the same lookup kinds as find --kind:

• stock
• rrid
• gene
• fbid
• component
• property
• property-exact
• driver-family
• relationship
• search
• auto

You can also stack explicit filter flags on export; multiple flags combine as AND:

• --stock
• --rrid
• --gene
• --component
• --fbid
• --property
• --property-exact
• --driver-family
• --relationship
• --search

Compound Find

find also subsumes compound filters. Default dataset: components.

Examples:

bdsc find --gene Or56a --property lexA
bdsc find --gene Or67d --property qf
bdsc find --gene Or42b --driver-family lexA
bdsc find --gene Or56a --property-exact lexA
bdsc find --dataset stocks --property optogenetic
bdsc find --dataset genes --property olfactory --relationship coding --jsonl

Reports

Use report for curated high-level buckets that would otherwise need multiple queries or OR filters.

Reports:

• olfactory: receptor-family genes (Or*, Orco, Ir*, Obp*)
• drivers: GAL4 / lexA / QF / split-driver / FLP-like driver surfaces
• optogenetics: common optogenetic effectors plus optogenetic-tagged properties

Examples:

bdsc report olfactory
bdsc report olfactory --dataset genes --jsonl
bdsc report drivers --limit 50 --json
bdsc report optogenetics --dataset components --jsonl

Terms

Use terms when you need to discover the vocabulary before filtering.

Scopes:

• properties
• property-descriptions
• relationships

Examples:

bdsc terms properties --limit 20
bdsc terms properties --query VALIUM --json
bdsc terms relationships --limit 20
bdsc terms property-descriptions --query optogenetic --jsonl

Notes

• sync uses conditional HTTP headers when possible (ETag, If-Modified-Since) to avoid re-downloading unchanged files.
• Local lookup is built from the public CSV dumps, not the private site search endpoints.
• search now uses a two-stage index: exact/prefix FTS first, trigram fuzzy fallback second. Typos and loose spacing/punctuation usually still find the intended stock without having the exact BDSC string.
• find is the intended interactive entrypoint; dedicated legacy query commands still work but are no longer the main documented path.
• direct lookup paths also rerank fuzzy candidates when exact/prefix matching misses.
• use property-exact or driver-family when property is too broad for a reliable LexA/QF/GAL4-style answer.
• tag pushes like vX.Y.Z run the release workflow: build artifacts, create a GitHub release, and publish to PyPI.
• scripts/render_homebrew_formula.py renders a Homebrew formula from a built sdist; use it when updating a tap after a release.
• The live endpoint is undocumented and may change without notice.
• BDSC data is large enough that the first full sync/index can take a few minutes depending on network and disk speed.