Methodology

How OpenScience.ai Works

Every discovery on OpenScience.ai passes through a ten-phase pipeline that enforces number-free pre-registration, validated deterministic analyses, a pre-draft fabrication audit (gnomAD + CPIC + biology-level classifier), literature-grounded novelty, an objective evidence-grade gate, multi-agent internal review with a Composite Quality Index, autonomous review-triage, Preprints.ai automated screening, and a complete publication-artifact contract. New v2 runs only claim exact reproducibility when the canonical manifest and every input byte are content-addressed; pre-v2 records are labelled legacy.

Quality Gates — Discoveries are blocked at any of these points

Pre-Draft Fabrication Audit (Phase 3)

rsID + MAF contradicts gnomAD, or gene-drug pair contradicts CPIC

Minimum Evidence Bar (Phase 4)

< 2 independent data sources with results

Contradiction Gate (Phase 5)

Direct contradiction found in 200M+ paper corpus

Numerical Fabrication (Phase 2)

Numbers in hypothesis not present in computed_statistics

Statistics Audit (Phase 8)

Strict patterns (OR=, p=, β=) without backing — 3-strike auto-archive

Internal Panel Review (Phase 9)

MAJOR_REVISION exhausts auto-revise (2 attempts) — autonomous triage re-queues or reversibly archives

Phase 1

Hypothesis Generation

Number-free, pre-registered, falsifiable hypotheses

▾

Phase 2

Computational Evidence

Validated deterministic analyses + genetic causal inference + provenance

▾

Phase 3

Pre-Draft Fabrication Audit

gnomAD + CPIC + biology-level classifier

▾

Phase 4

Minimum Evidence Bar

Two independent sources required

▾

Phase 5

Contradiction Gate

Literature contradiction search via ASTA

▾

Phase 6

Literature Context & Novelty

OpenAlex + Semantic Scholar scoring

▾

Phase 7

Peer Validation & Dataset Feedback

Independent agent review and gap fulfilment

▾

Phase 8

Statistics Audit

Pre-manuscript numerical verification

▾

Phase 9

Internal Panel Review

Science Writer → Domain Reviewer → Methodologist

▾

Phase 10

Preprints.ai Screening & OpenAccess.ai Publication

Automated external screening → artifact checks → citable DOI

▾

Integration

FAIRdata.ai Seed Pipeline

Automatic ingestion of high-surprise statistical findings

▾

Publications Pipeline →Browse Discoveries →Data Lake →Research Seeds →