A single, searchable home for publicly available QTL data. locusview aggregates published quantitative trait locus (QTL) data — bulk-tissue and single-cell — and lets researchers search, browse, and download it from one place, instead of hunting across a dozen scattered portals.
Status: Phase 0 — Foundation. We are building the project (repository, process, docs, and product definition) before building product features. There is no working portal yet. See the roadmap.
This repository is also a teaching artifact: it is being built, in the open, as the worked example for a graduate course on software engineering and AI-agent-native development. If you are a student, start with the software-engineering lifecycle explainer.
Full, verified steps live in
docs/how-to/. The short version:
# 1. Install uv (https://docs.astral.sh/uv/) if you don't have it, then:
uv sync # create the virtual env and install dependencies
uv run pytest # run the test suite
uv run locusview # run the app (a placeholder until Phase 1)You will also need the genomics toolchain (bcftools, tabix, bgzip from HTSlib) for Phase 1;
the machine-setup how-to covers a known-good path (a container is recommended).
We organize docs with Diátaxis. See docs/README.md.
| I want to… | Go to |
|---|---|
| Understand why we work this way | docs/explanation/ |
| Do a specific task | docs/how-to/ |
| Look up a fact (schema, API) | docs/reference/ |
| Learn by building (student labs) | docs/tutorials/ |
| See product intent (Vision, PRD, roadmap) | docs/product/ |
| See how we work & use agents | docs/process/ |
| See decisions and why | docs/adr/ |
| See the course layer | docs/course/ |
See CONTRIBUTING.md. In short: small feature branches, pull requests with green
CI, human review before merge, and agents write, humans merge.
MIT — open source from commit #1, so students can follow the whole history.