Patient Cohorts
A biomedical data marketplace connecting hospitals and labs with pharma and biotech researchers, unlocking value from patient-level genomic and clinical data through a Trusted Research Environment where raw data never leaves the provider's infrastructure.
The Problem
The data exists. The connection doesn't.
Hospitals sit on rich genomic and clinical datasets. Pharma teams can't access them. Patients from underrepresented populations are excluded from the research that shapes their treatments.
Data sitting dormant
Hospitals and diagnostic labs generate massive volumes of genomic and clinical data. Once the report is delivered, that data becomes a dormant cost center, stored, maintained, but never monetized.
Public databases skew European
Meanwhile, pharma teams struggle to find real-world patient data from non-European populations. Public databases heavily skew toward European ancestry, leaving MENA, South Asian, and African genomic diversity severely underrepresented.
No compliant bridge exists
There's no scalable, compliant way to connect hospitals that have the data with researchers who need it. The result: data stays locked in silos, and drug development proceeds on incomplete evidence.
How It Works
Two sides. One marketplace.
Data providers list cohorts. Researchers discover and license access. Raw data never changes hands, all analysis happens in a Trusted Research Environment.
Step 01
Pan.bio inventories your existing data assets, genomic, clinical, imaging, proteomic, and phenotype annotations, to understand what's available.
What each side gains from the marketplace
New revenue stream
Monetize existing data, no additional sequencing cost required
Full data ownership
Data stays under your control and jurisdiction at every step
Compliance by architecture
In-country processing and TRE model handle the regulatory hard parts
Recurring revenue
The same cohort can be licensed to multiple researchers simultaneously
Zero upfront cost
Pan.bio handles platform infrastructure and researcher acquisition
For Hospitals & Labs
- Data Assessment: Pan.bio inventories your existing data assets, genomic, clinical, imaging, proteomic, and phenotype annotations, to understand what's available and how it can be structured.
- Structuring & Standardization: Data is mapped to Pan.bio's schema using recognized ontologies (HPO, ICD-10, LOINC) and standard genomic formats, ensuring consistency across institutions.
- De-Identification: Patient data is de-identified per HIPAA and local regulations. All processing happens locally, data doesn't leave the institution during this phase.
- Listing: The cohort is published on the marketplace with metadata, disease focus, data modalities, patient count, and completeness score, visible to researchers.
- Revenue: When researchers license access, the provider earns a majority share of the fee on every access event. The same cohort can be licensed to multiple researchers simultaneously.
Hospital Benefits
- New revenue stream: Monetize existing data, no additional sequencing cost required
- Full data ownership: Data stays under your control and jurisdiction at every step
- Compliance by architecture: In-country processing and TRE model handle the regulatory hard parts
- Recurring revenue: The same cohort can be licensed to multiple researchers simultaneously
- Zero upfront cost: Pan.bio handles platform infrastructure and researcher acquisition
For Pharma & Biotech
- Discovery: Search the marketplace by disease, data modality, population, or genomic criteria. Cohort metadata is visible before any commitment.
- Feasibility: Review aggregate summaries, demographics, variant frequencies, and completeness scores to assess fit against your research question.
- Licensing: Execute a license specifying duration, scope, and pricing. Pricing is per-patient, per-year, segmented by data type.
- Workspace: Pan.bio provisions an isolated Trusted Research Environment pre-loaded with the licensed cohort and analytical tools.
- Analysis & Export: Run analyses within the workspace. Export results and derived insights, never the raw data.
Researcher Benefits
- Underrepresented populations: Access cohorts unavailable in public databases, particularly MENA and South Asia
- Weeks, not months: No more spending a quarter chasing data access agreements
- Multi-modal linkage: Genomics, clinical, imaging, and proteomics linked at the individual patient level
- Reproducible environment: Secure, compliant compute environment with pre-loaded tools
- Data freshness: Active clinical institutions, not static snapshots published years ago
What's in a Cohort
Multi-modal data. Linked at the patient.
Unlike single-purpose databases, Pan.bio cohorts link genomics, clinical records, and imaging at the individual patient level, giving researchers the complete picture.
Genomics
Clinical
Imaging
Patient Metadata
Demographics, ancestry, consent flags, and longitudinal follow-up data, enabling precise cohort filtering and reproducible study design.
Find the right cohort.
Ask, don't filter.
BioMind powers two specialized agents within the cohort marketplace, translating research questions into cohort matches and surfacing gaps before you commit.
BioMind
AI Your AI research partner
for cohort discovery.
BioMind agents work at two levels: the Discovery Agent translates your natural-language research question into structured cohort filters with transparent gap analysis, while the Research Agent reviews available cohort metadata against your objectives and provides comparative recommendations.
Cohort Discovery Agent
Translates natural-language queries into structured filters and surfaces matching cohorts with transparent gap disclosure, ranking the best available options and clearly showing where datasets fall short.
Cohort Research Agent
Reviews the research objective against available cohort metadata and provides comparative recommendations, highlighting strengths, limitations, and potential gaps of each candidate dataset.
Transparent Gap Disclosure
When no perfect cohort exists, the agents explicitly disclose what's missing, partial ancestry coverage, limited imaging modality, incomplete follow-up, so researchers make informed decisions, not blind ones.
Not another data broker.
A fundamentally different model.
Population Diversity
Public databases skew heavily European-ancestry. Pan.bio targets underrepresented populations, particularly MENA and South Asia, with data that doesn't exist elsewhere.
Controlled Access by Architecture
Data never leaves the provider. Researchers work in a Trusted Research Environment, fundamentally different from data transfer or download models.
Multi-Modal Linkage
Unlike single-purpose databases, Pan.bio links genomics, clinical records, imaging, and proteomics at the individual patient level.
Revenue Sharing That Aligns Incentives
Providers earn on every access event. Public databases don't compensate data originators, and traditional brokers often offer unfavorable terms.
Enforced at the infrastructure level, your data stays in your jurisdiction, always.
Ready to Connect?
Whether you're a hospital looking to unlock value from existing data or a researcher seeking diverse, real-world cohorts, Pan.bio provides the marketplace and infrastructure to make it happen.
No credit card required · Start in minutes