Big data for better health
Olfactory function predicts brain health, disease risk, and overall well-being years before symptoms emerge, for example Alzheimer’s.
Yet it remains one of the least measured vital signs in clinical practice.
Isolated datasets, inconsistent protocols, and fragmented infrastructure have kept olfactory science from reaching its full potential.
Connecting existing data across sites and cohorts is the single most powerful step we can take to accelerate scientific discovery and improve patient outcomes.
Hub4Smell makes the connection
Hub4Smell platform is a commercial innovation built specifically for the olfactory science and clinical community. It provides:
1. A next-generation data repository: a secure, AI-powered platform for storing, curating, and analyzing human smell test data, built with tools designed for olfactory datasets specifically, unlocking smell test data to solve problems that no single study can address alone.
I wish I could...
- ...do instant benchmarking of my own data
- ...compare smell scores across testing tools and samples easily
- ...see large scale longitudinal data on the broad population
- ...easily choose which smell test is best for my purpose
- ...access harmonized data
Every dataset contributed brings us measurably closer to making these wishes a reality.
2. A collaborative community: a professional network where researchers and clinicians exchange knowledge, co-design impactful studies, and advance universal smell testing as a standard of care.
AlphaFold: From Data Scarcity to Discovery Acceleration
In several scientific fields, progress remained incremental until data reached critical mass. Structural biology is a striking example: once large, standardized protein datasets enabled tools like AlphaFold, reliable 3D structure prediction from sequence became routine, yielding models for hundreds of millions of proteins and sharply accelerating structural and mechanistic biology.
When data are harmonized and aggregated, their value multiplies. Patterns emerge that are invisible within single studies. Longitudinal signals become detectable. Predictive models become feasible. Clinical translation accelerates.
The Hub4Smell platform is designed to catalyze this shift.
Join others to be a catalyst for change of smell for health.
Contribute data and knowledge to Hub4Smell.
Share your KNOWLEDGE.Hub4Smell amps your impact.
As data are harmonized and aggregated, their value multiplies. When expertise is shared, not siloed, methods, insights and connections are amplified. In an environment where both data and expertise circulate freely, progress accelerates exponentially.
Join the community and leverage the energy of a single purpose to strengthen the evidence of olfaction’s meaning to health
Who has access?Who benefits from you sharing?
Your data is more powerful and informative when placed into context. The infrastructure offered by the Hub4Smell platform strives to provide connected datasets, consistent protocols, and integrated infrastructure. It delivers a FAIR architecture where data can be found, accessed appropriately, integrated across studies, and reused with responsible protections in place.
- All anonymous data is available by Conversational AI for deep insights
- Contributors retain control over who can download their own dataset
- All access requests are reviewed and logged for full transparency
What’s to come...
Smell Data. Shared Insight. Universal Impact.
Before any data enters the Hub4Smell repository, it must be anonymized so that no individual participant can be re-identified by any available method. This is required for comprehensive data privacy compliance.
Quick Anonymization checklist — remove before sharing:
- IDs, such as names, numbers, addresses, other identifiers
- Keys or existing list linking names to data
- Dates, such as month or day (of birth or testing date)
- Specific location information (only country-level information or zip codes)
- Special circumstances, such as rare diseases or high-dimension omics (e.g., sequenced genomes)
Share your data in 5 steps:
- 1
Register: Create a contributor account at beta.hub4smell.org using your institutional email.
- 2
Anonymize: Apply the checklist above as a starting point. The End User License Agreement and Privacy Policy have full details.
- 3
Confirm ethics coverage: Verify that your original consent covers secondary data use, or that your data qualifies as fully anonymized. Our team can advise if you are unsure.
- 4
Upload: Using the Hub4Smell data template, map your variables to our standard schema, and upload securely via the contributor portal.
- 5
Start exploring: Your dataset is live in the repository with your attribution. Use Hub4Smell's tools to benchmark, compare, and test your hypotheses immediately.