build: pixi.lock, conda environment.yml, Dockerfile, GHCR release workflow

.dockerignore

@@ -0,0 +1,11 @@
+# Only pixi.toml, pixi.lock, pyproject.toml, README.md, src/ and db/ go into the image.
+*
+!pixi.toml
+!pixi.lock
+!pyproject.toml
+!README.md
+!src/
+!db/
+**/__pycache__
+db/*.duckdb
+db/candidates/

.github/workflows/docker.yml

@@ -0,0 +1,83 @@
+name: docker
+
+# Release images: ghcr.io/stajichlab/matpredict:<version> and :latest.
+# A v* tag or a published release builds and pushes; a pull request that touches
+# the image inputs builds and smoke-tests without pushing; workflow_dispatch
+# builds and pushes on demand.
+on:
+  push:
+    tags: ["v*"]
+  release:
+    types: [published]
+  workflow_dispatch:
+  pull_request:
+    paths:
+      - Dockerfile
+      - .dockerignore
+      - pixi.toml
+      - pixi.lock
+      - pyproject.toml
+      - .github/workflows/docker.yml
+
+permissions:
+  contents: read
+  packages: write
+
+env:
+  IMAGE: ghcr.io/${{ github.repository_owner }}/matpredict
+
+jobs:
+  image:
+    runs-on: ubuntu-latest
+    steps:
+      - uses: actions/checkout@v4
+
+      - name: Image tags
+        id: meta
+        uses: docker/metadata-action@v5
+        with:
+          images: ${{ env.IMAGE }}
+          tags: |
+            type=semver,pattern={{version}}
+            type=semver,pattern={{major}}.{{minor}}
+            type=raw,value=latest,enable=${{ startsWith(github.ref, 'refs/tags/v') || github.event_name == 'release' }}
+            type=ref,event=pr
+            type=sha,format=short
+
+      - uses: docker/setup-buildx-action@v3
+
+      - name: Build (load locally for the smoke test)
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          load: true
+          tags: matpredict:test
+          build-args: VERSION=${{ steps.meta.outputs.version }}
+          cache-from: type=gha
+          cache-to: type=gha,mode=max
+
+      - name: Smoke test
+        run: |
+          docker run --rm matpredict:test --version
+          docker run --rm matpredict:test detect --help > /dev/null
+          docker run --rm --entrypoint /bin/bash matpredict:test /shell-hook.sh \
+            bash -c 'for t in tblastn diamond miniprot exonerate mafft augustus taxonkit; do command -v $t > /dev/null || { echo "missing $t"; exit 1; }; done; python -c "import pyhmmer, MATPredict"'
+
+      - name: Log in to GHCR
+        if: github.event_name != 'pull_request'
+        uses: docker/login-action@v3
+        with:
+          registry: ghcr.io
+          username: ${{ github.actor }}
+          password: ${{ secrets.GITHUB_TOKEN }}
+
+      - name: Push
+        if: github.event_name != 'pull_request'
+        uses: docker/build-push-action@v6
+        with:
+          context: .
+          push: true
+          tags: ${{ steps.meta.outputs.tags }}
+          labels: ${{ steps.meta.outputs.labels }}
+          build-args: VERSION=${{ steps.meta.outputs.version }}
+          cache-from: type=gha

.gitignore

@@ -3,7 +3,6 @@
 .matpredict_cache/
 __pycache__/
 *.pyc
-pixi.lock
 .pytest_cache/
 .pixi/
 db/*.duckdb

CHANGELOG.md

@@ -4,6 +4,37 @@ All notable changes to MATPredict. Versions follow [semantic versioning](https:/
 release tags may carry a clade suffix naming the lineage whose infrastructure that
 release completed.
 
+## [Unreleased]
+
+### Changed
+- No default NCBI e-mail. The address comes from `$MATPREDICT_NCBI_EMAIL` or the
+  `[ncbi]` table of `~/.config/matpredict/config.toml` (`$MATPREDICT_CONFIG`,
+  `$XDG_CONFIG_HOME`). Without one, requests carry no e-mail and one warning is
+  logged. Requests now send `tool=MATPredict`.
+- The NCBI response cache key leaves out `email`, `api_key` and `tool`. Entries
+  cached under the old full-URL key are still found and copied forward.
+- README rewritten: goals, workflow, curation process, quickstart, usage,
+  validation, citation.
+
+### Added
+- `pixi.lock` is now committed, so environments and images are reproducible.
+- `environment.yml` for conda/mamba users (exported from `pixi.toml`).
+- `Dockerfile` (pixi build stage, Ubuntu 24.04 runtime with the environment,
+  the package and `db/`).
+- `.github/workflows/docker.yml`: builds and smoke-tests the image on pull
+  requests that touch the image inputs; on a `v*` tag, a published release or a
+  manual run it also pushes `ghcr.io/stajichlab/matpredict:<version>` and
+  `:latest`.
+
+## [0.6.0] — 2026-10-03 — `v0.6.0`
+
+Post-#9 work merged as PR #10: confidence and tier rules, the V3 polish cap,
+the flank-carried rule, `not_searched` routing, the MAT-gene gate, the P1
+paralog class, deterministic classifier builds, scope-only families,
+`idiomorph_class`, the regression check, new curated records, and held-out
+validation (LCG 536/621, Jena 61/64, Zygo 23/23). See the `v0.6.0` tag message
+and `analysis/INDEX.md`.
+
 ## [0.5.0] — 2026-09-20 — `v0.5.0-mucoromycota`
 
 Mucoromycota detection becomes usable end to end: idiomorph calling works, the

Dockerfile

@@ -0,0 +1,34 @@
+# MATPredict container image.
+# Build stage: solve nothing, install the locked pixi environment (pixi.lock).
+# Runtime stage: the environment, the package source and the curated database.
+#
+#   docker build -t matpredict .
+#   docker run --rm -v "$PWD":/data -e MATPREDICT_NCBI_EMAIL=you@example.org \
+#     matpredict detect --genome /data/genome.fna --taxid 4837 --out-dir /data/out
+
+FROM ghcr.io/prefix-dev/pixi:0.71.3-noble AS build
+WORKDIR /app
+COPY pixi.toml pixi.lock pyproject.toml README.md ./
+COPY src ./src
+RUN pixi install --locked -e default \
+ && pixi shell-hook -e default -s bash > /shell-hook.sh \
+ && echo 'exec "$@"' >> /shell-hook.sh
+
+FROM ubuntu:24.04 AS runtime
+ARG VERSION=dev
+LABEL org.opencontainers.image.title="MATPredict" \
+      org.opencontainers.image.description="Find and type fungal mating-type (MAT) loci in genome assemblies" \
+      org.opencontainers.image.source="https://github.com/stajichlab/MATPredict" \
+      org.opencontainers.image.version="${VERSION}"
+# The environment keeps its build path (/app/.pixi/envs/default): conda
+# packages and the editable MATPredict install record absolute paths.
+COPY --from=build /app/.pixi/envs/default /app/.pixi/envs/default
+COPY --from=build /shell-hook.sh /shell-hook.sh
+COPY --from=build /app/src /app/src
+COPY pyproject.toml README.md /app/
+COPY db /app/db
+ENV MATPREDICT_DB_ROOT=/app/db \
+    MATPREDICT_CACHE_DIR=/data/.matpredict_cache
+WORKDIR /data
+ENTRYPOINT ["/bin/bash", "/shell-hook.sh", "matpredict"]
+CMD ["--help"]