Catalogue
Discovery in place
Crawls your platforms where they are, profiles and classifies every column for PII and PHI, and scores quality.
Governed Life Sciences Data & AI Intelligence Platform
An intelligence layer above every platform your data already lives on.
CYRENYX discovers, catalogues, contextualises and governs Life Sciences data across clinical, imaging, real-world, omics and document systems, and gives people and AI agents one governed way to find, trust and use it. Your data stays where it lives, and every deployment is customised to your own architecture.
Product demo · 90 seconds
From a CRO delivery to governed, ready-to-query data
A study delivery is onboarded while you watch: agents tag every column, score its quality, link it into the knowledge graph and apply the right privacy controls before anyone queries it.
The problem
Questions teams still can't answer about their own data
Life Sciences data is spread across clinical systems, imaging platforms, real-world data, omics, document stores, lakehouses and warehouses. Every one of these questions is really about relationships, and CYRENYX keeps those relationships as a governed graph.
- What data exists, and where does the authoritative copy live?
- Did another team already build this exact dataset?
- How was it transformed to get here, column by column?
- Does it contain PII or PHI, and what consent or reuse rules apply?
- Am I allowed to use it for this purpose, and can I reuse it scientifically?
- Which analyses, dashboards and submissions depend on it?
The platform
One governed layer, not another warehouse
CYRENYX doesn't replace Databricks, Snowflake or your cloud. It sits across them, reads only metadata and the samples you permit, and turns scattered storage into knowledge people and agents can trust.
People & agents
ResearchersData stewardsStudy teamsAI agentsRegulatory & audit
CYRENYXdiscovery · ontology · lineage · policy · conversational search
DatabricksSnowflakeAWS · Azure · GCPVeevaMedidataPACS / DICOMLIMS · ELNeTMF
Source platforms · data stays here
Ontology
Knowledge graph
A Life Sciences model that links datasets to studies, molecules, endpoints, consent and the analyses built on them.
Lineage
Column by column
How every column got here, which copy is the source of truth, and what breaks downstream if it changes.
AI
Governed agents
Ask in plain language. Read-only agents answer with citations, and every step they take passes an access check.
Onboarding
Deliveries, protected
CRO and vendor deliveries land with pseudonymisation or encryption applied per data type, linked in a patient registry.
Workbench
Analysis on governed data
Notebooks, pipelines and dataset workspaces that only ever see what the policy engine allows, with approved libraries.
Governance
Every query is decided by policy
The same policy engine decides what each person and each agent sees, as full access, masked, aggregate-only or metadata-only, for the purpose they declared.
Purpose of use
Every session declares a purpose. Change the purpose, and access changes with it, live.
Consent & reuse
Consent bases and reuse restrictions travel with the data, and small cells are suppressed in aggregates.
Tamper-evident audit
Every query, tool call and agent step is written to a hash-chained audit trail that can be verified at any time.
Secrets by reference
Credentials never land in the catalogue, the logs or the audit trail, and agents can read but never change data.
Live demo
Try it on synthetic data
The live demo runs Demo Pharma, a fictional sponsor with oncology, immunology and neuroscience studies. Sign in as a scientist, a data owner or a steward and watch access change with the role and the purpose.
- Sponsor
- Demo Pharma (fictional)
- Studies
- DMO-2023-014, DMO-2024-001 and more
- Data
- Clinical, imaging, omics, registry and documents, all synthetic
- Access
- By invitation; personas from scientist to data owner
Customisable
Fitted to your architecture, not the other way round
No two Life Sciences landscapes look alike. Every CYRENYX deployment is customised to the client's own architecture: where it runs, what it connects to, who signs in and which rules apply.
Deployment
Runs where you choose
In your own AWS, Azure or Google Cloud account, or on your Kubernetes cluster, set up with infrastructure as code.
Integration
Connects to your platforms
Over 25 ready connectors, from Veeva Vault, Medidata Rave and SharePoint to DICOMweb, FHIR and S3, plus new ones for your own systems.
Identity
Your sign-in
Single sign-on with Microsoft Entra ID, AD FS, Okta or any SAML 2.0 provider, with your groups mapped to roles.
Governance
Your rules
Purposes of use, access policies, consent bases and approval workflows modelled on how your organisation already works.
Organisation
Your structure
Affiliates, CROs and partners as separate tenants, each with its own data residency, encryption keys and licensed modules.
Context & AI
Your ontology and models
Extend the ontology with your own entities and terms, and use Claude through your own account or run without an external model.
Talk to Astatinedata about CYRENYX
Questions, a walkthrough for your team, or a pilot fitted to your own architecture: tell us what you're exploring.