BioResearch Assistant
Literature mining, pseudonymisation, MII export — on-premise by default, in your infrastructure. Optional external LLM APIs only if you configure them.
Early accessWhy it exists
Clinical research data usually cannot simply go to the cloud. Existing tools ignored that — or required complicated workarounds. BioResearch Assistant is built for exactly this case: pseudonymisation at the data holder, interoperable interfaces when needed, full control over your own infrastructure.
BioResearch Assistant combines literature mining, privacy-conscious pseudonymisation and MII-oriented export — for on-premise use in research and clinical settings. On-premise by default with local LLM options (typically Ollama). External API providers — including Anthropic Claude if the operator supplies an API key — are optional and configurable; those calls then go to that provider under the operator’s account, not a Synaptic Four-hosted demo. GA4GH-oriented interfaces are ready when you need to connect to broader research infrastructure. Optionally, the Locus module adds a fully locally operated RAG stack for clinical bioinformatics.
Features
Literature Mining
Pseudonymisation (privacy-conscious)
MII Export (FHIR bundles)
Consent Tracker (FHIR Consent)
Local GA4GH DRS/WES (optional Ferrum proxy) + Phenopackets
Async export jobs + validation reports
Locus — on-premise RAG (optional, curated indexes)
Optional module — implemented (beta)
Locus — on-premise RAG for clinical bioinformatics
Semantic literature search and structured queries (RAG architecture, on-premise). API and database layer are implemented; curated index delivery and governance review remain engagement-specific. LOCUS-MODULE documentation is in the public repository.
LOCUS-MODULE documentation on GitHub →Orientation, not legal advice
Technical foundation for on-premise research workflows. Legal compliance (e.g. GDPR) stays with you as operator — not a certification and not automatic compliance.
Pricing & getting started
| Pilot (first customer) | Free or symbolic — we want real feedback, not an early invoice. |
| Licence (from second customer) | €3,500–4,000/year — all-in, no separate one-time fee. |
| Onboarding & setup | €2,500 — one-time, together with you. |
| Support | €600/month — for teams that want ongoing accompaniment. |
Indicative figures for the BioResearch Assistant product. Broader institutional Synaptic Research AI programmes (multi-faculty intelligence infrastructure) are scoped separately — ask if that is your path. Binding terms in writing.
For a whole faculty or institution: talk to us directly.
What it is not
- Not a cloud service — operates in your own infrastructure.
- No formal certification (ISO, MDR, HIPAA).
- No automatic compliance — responsibility lies with the operator.
- Local DRS/WES are BRA’s own surfaces. Set FERRUM_DRS_URL / FERRUM_WES_URL to proxy those APIs to Ferrum — BRA is not the institutional archive.
- There is no hosted public BRA demo. Clone and run locally. Optional cloud LLM keys are the operator’s processing, not a Synaptic Four SaaS.
Licensing
Business Source License 1.1 – academic and other permitted uses per LICENSE; commercial production use requires a licence. Converts to Apache-2.0 after four years. The public repository includes the full licence text.
When you outgrow a team tool
BRA is a departmental research product. Institutional GA4GH production APIs are Ferrum’s lane — different buyer, different price band on purpose.
Regulatory context: EHDS · GDPR & health data
A straight assessment of whether BioResearch Assistant matches your infrastructure—without a forced cloud model.