Aquaculture Resilience Evidence Engine (AREE)

AREE is an open, reproducible research resource that converts fragmented public omics datasets into harmonized, comparable resilience-biomarker evidence for Pacific oyster (Crassostrea gigas) and, by design, other shellfish and aquaculture organisms.

It is not a static resource list or a generic website. It is a versioned evidence-generation system with five connected layers:

  1. Study registry & intake — machine-readable registration of public datasets with controlled vocabularies for phenotypes and stressors.
  2. Standardized reanalysis workflows — Nextflow workflows for RNA-seq, methylation, proteomics, and metabolomics (raw-reanalysis or processed-results modes).
  3. Cross-study harmonization — a shared evidence schema plus identifier mapping with explicit confidence levels.
  4. Meta-analysis & candidate prioritization — random-effects pooling, heterogeneity statistics, and a transparent (non-black-box) candidate score with hard tier gates.
  5. User-facing outputs — this documentation site plus a Streamlit interface and downloadable evidence cards.

Core principles

  • Association, not proof. AREE surfaces evidence convergence, never confirmed mechanistic causation. A statistically significant hit in one study is never presented as a validated biomarker.
  • Provenance always. Every result traces back to a source file, checksum, parameter set, and workflow version.
  • Missingness is explicit. Missing metadata is recorded and reportable, never silently dropped or imputed.
  • Contradictions are surfaced. Conflicting cross-study findings appear in meta-analysis summaries and evidence cards rather than being averaged away.

Where to go next

Note on demo data. Every demo dataset is labeled simulated and is partitioned from real evidence during pooling, ranking, and reporting. The repository also includes registered real public studies; do not cite a demo result as real biological evidence.