Skip to content

Installation and Setup

Install bijux-proteomics-knowledge to model scientific evidence and claims, normalize evidence into reviewable memory, preserve contradictions, resolve biological identities and associations, measure annotation coverage, and assemble decision briefs.

Requirements

  • Python 3.11 or newer
  • an isolated Python environment
  • compatible Foundation and Core packages, installed automatically
  • explicit source material for any knowledge being curated

The package does not download reference databases, literature, ontologies, or credentials during installation. Bundled fixtures support reproducible tests; they are not substitutes for current external scientific sources.

Install from PyPI

python -m venv .venv
source .venv/bin/activate
python -m pip install --upgrade pip
python -m pip install bijux-proteomics-knowledge

Confirm schema compatibility through the public surface:

from bijux_proteomics_foundation import DocumentSchema
from bijux_proteomics_knowledge import evaluate_schema_compatibility

schema = DocumentSchema(
    created_by="knowledge-setup-check",
    document_kind="annotation_pack",
    package_name="bijux-proteomics-knowledge",
)
report = evaluate_schema_compatibility(schema)

assert report.compatible

That check establishes document compatibility only. It does not establish that annotations are current, identifiers are resolvable, citations support a claim, or a bundle is free from contradiction.

Prepare a reviewable evidence fixture

Include enough variation to exercise the package contract:

  • unique evidence and claim identifiers;
  • source type, origin, extraction method, dates, and citations;
  • organism and scientific context where identity depends on them;
  • quantitative support with units and uncertainty when applicable;
  • duplicate, malformed, stale, ambiguous, and contradictory cases;
  • expected ingestion counts and rejection reasons;
  • expected graph-integrity and reconciliation outcomes.

Separate external references from checked-in fixture records in manifests and descriptions. A reproducible fixture proves behavior against a known case; it does not prove external completeness or freshness.

Source checkout

From the repository root:

python -m pip install -e "packages/bijux-proteomics-knowledge[test]"
python -m pytest packages/bijux-proteomics-knowledge/tests

During development, run the matching families under tests/contracts, tests/memory, tests/references, tests/reviews, and the biological resolver tests. A curation path is ready only when normalization, provenance, ambiguity, contradiction, coverage, and deterministic rendering are all verified.