Conversation
- Add 'migrate' one-shot service that runs alembic upgrade head before API starts - Mount docker/init.sql to postgres initdb.d to enable pgvector extension automatically - All workers and API depend on migrate completing successfully
- Add docker-compose.prod.yml: uses pre-built ghcr.io images, GPU support for worker-embeddings - Rewrite deploy_vast.sh: sync compose files + docker compose pull + up (no rsync of source code) - Migrations run automatically via the migrate service on every deploy
- ScoringConfig: reproducible scoring recipe with signal weights, formula type and optional per-evidence-code quality overrides - EvaluationResult: add scoring_config_id FK and results JSONB column - OntologySnapshot: add ia_url field for Information Accretion file - SupportEntry: new model for user-facing support/contact entries - 6 new Alembic migrations covering all model additions and indexes Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- scoring.py: configurable multi-signal scoring engine combining embedding similarity, NW/SW alignment identity, taxonomic proximity and per-evidence-code quality weights; supports weighted_avg and evidence_weighted formulas - metrics.py: shared metric computation utilities - evaluation.py: propagate scoring_config through evaluation runs, store per-namespace Fmax/precision/recall/coverage in results JSONB - run_cafa_evaluation: download IA file from OntologySnapshot.ia_url, apply scoring config to generate CAFA-format prediction scores - predict_go_terms: wire scoring_config_id into prediction batch payload Co-Authored-By: Claude Sonnet 4.6 <noreply@anthropic.com>
- scoring.py: CRUD for ScoringConfig + preset factory + scored TSV
streaming endpoint (/scoring/prediction-sets/{id}/score.tsv)
- support.py: contact/support entry submission endpoint
- annotations.py: evaluation-set DELETE cascade, result DELETE,
artifacts ZIP download, IA-url PATCH on snapshots, ground-truth
FASTA and TSV download endpoints
- embeddings.py: propagate scoring_config_id in prediction launch
- app.py: register scoring and support routers
- base_worker.py: surface scoring_config_id in worker dispatch
…verhaul - New /scoring page: create/delete ScoringConfig with signal weights, evidence overrides, formula selector and preset loader - New /support page: contact form backed by support router - SupportButton + UsagePolicyModal components - evaluation/page.tsx: full rewrite — per-result metrics table (NK/LK/PK × BPO/MFO/CCO), scoring config selector, artifact ZIP download, result polling after job submit, IA-url management on snapshots - functional-annotation: scoring config selector in predict form, enriched per-protein prediction detail view - annotations/page.tsx: IA-url PATCH UI on ontology snapshots - NavLinks: add Scoring and Support entries - lib/api.ts: ScoringConfig CRUD, scored TSV URL helper, IA-url PATCH, evaluation result DELETE and artifact endpoints - fix: NaN in limit-per-entry input (guard parseInt) - fix: DELETE 204 response parsed as JSON (use fetch directly)
- evaluation.rst: full CAFA evaluation protocol documentation (NK/LK/PK categories, scoring config integration, data model) - operations.rst: document run_cafa_evaluation and scoring engine - index.rst, core.rst: include new modules in autodoc - .gitignore: exclude data/ref_cache/ (large .npy files) and test-results chore: add LICENSE, IA benchmark data, Playwright e2e scaffolding
Add test_metrics.py (15 tests) covering PRPoint, CAFAMetrics.summary, compute_cafa_metrics validation and correctness to raise test coverage above the 65% CI threshold.
Add 5 new test files covering: - scoring engine (evidence_weight, compute_score, score_predictions) - CAFA evaluation dataclass and _get_descendants BFS - GenerateEvaluationSetOperation payload validation and execute path - scoring CRUD router (configs, presets, 404 preflight checks) - support and maintenance routers (GET/POST support, vacuum sequences/embeddings) Extend test_core.py with normalize, is_experimental and RetryLaterError tests. Total: 399 passing, 3 skipped.
Join EmbeddingConfig, AnnotationSet and OntologySnapshot in the list_prediction_sets query so each row returns model_name, annotation source+version and OBO version instead of raw UUIDs. Frontend consumes the new optional fields and falls back to shortId when they are absent.
Workers are now started with setsid so each gets its own process group. Stop/force-kill targets the whole group (kill -- -PID) to prevent orphaned child processes. Grace period reduced from 60s to 5s.
Clean up unused imports and unsorted import blocks flagged by ruff in scoring.py, metrics.py, core/scoring.py and orm models __init__.
Run ruff format on all Python and frontend package files. No logic changes — pure whitespace, line-length and import ordering.
Codecov Report❌ Patch coverage is Additional details and impacted files@@ Coverage Diff @@
## main #6 +/- ##
==========================================
- Coverage 65.81% 65.01% -0.81%
==========================================
Files 49 55 +6
Lines 3873 4550 +677
==========================================
+ Hits 2549 2958 +409
- Misses 1324 1592 +268 ☔ View full report in Codecov by Sentry. 🚀 New features to boost your workflow:
|
- Migrate all pages from app/ to app/[locale]/ (next-intl App Router) - Add middleware, i18n/routing.ts, i18n/request.ts for locale detection - Add LanguageSwitcher component with locale toggle - Add message files for en, es, de, pt, zh (~21–23 KB each) - Translate all hardcoded UI strings across 14 pages + 6 components: proteins, proteinDetail, annotations, embeddings, functionalAnnotation, evaluation, scoring, querySets, jobs, jobDetail, maintenance, support, statusBadge, eventTimeline, resetDbButton, supportButton, usagePolicyModal - Remove dead app/layout.tsx and old app/ page stubs - README: clarify Docker not yet validated; mark source install as recommended
This file contains hidden or bidirectional Unicode text that may be interpreted or compiled differently than what appears below. To review, open the file in an editor that reveals hidden Unicode characters.
Learn more about bidirectional Unicode characters
Sign up for free
to join this conversation on GitHub.
Already have an account?
Sign in to comment
1 participant
Add this suggestion to a batch that can be applied as a single commit.This suggestion is invalid because no changes were made to the code.Suggestions cannot be applied while the pull request is closed.Suggestions cannot be applied while viewing a subset of changes.Only one suggestion per line can be applied in a batch.Add this suggestion to a batch that can be applied as a single commit.Applying suggestions on deleted lines is not supported.You must change the existing code in this line in order to create a valid suggestion.Outdated suggestions cannot be applied.This suggestion has been applied or marked resolved.Suggestions cannot be applied from pending reviews.Suggestions cannot be applied on multi-line comments.Suggestions cannot be applied while the pull request is queued to merge.Suggestion cannot be applied right now. Please check back later.
Summary
ScoringConfigmodel +/scoringrouter)/scoring(scoring config CRUD + presets) and/support(contact form)ScoringConfig,SupportEntry,ia_urlonOntologySnapshot, 6 new Alembic migrationsCommits
feat(orm): ScoringConfig, SupportEntry + migrationsfeat(core): scoring engine + CAFA evaluation pipelinefeat(api): scoring and support routersfeat(frontend): scoring UI, support page, evaluation overhauldocs: evaluation architecture + .gitignore ref_cachechore: bump version to 0.2.0Test plan
poetry run pytestpassesalembic upgrade headapplies cleanly