Pheno.AI Open Innovation Program
TIMELINE
Applications:
​
Dec 1, 2025 - Feb 1, 2026
Notifications & onboarding:
​
March - April 2026
Review:
​
February 2026
Pheno.ai’s HPP Open Innovation Program enables academic researchers to work with the world’s most comprehensive deep-phenotype datasets. The program provides academic, non-commercial access to the Human Phenotype Project datasets, spanning multi-omics, clinical measures, real-world lifestyle data, continuous glucose monitoring and multi-night sleep studies gathered at unparalleled scale and fidelity.
By opening this resource to the academic community, the program accelerates discovery, supports rigorous data-driven science and enables new insights into human biology. It fosters collaborations that push the boundaries of what can be learned from high-resolution human data - offering researchers a rare opportunity to translate unprecedented depth and breadth of information into meaningful scientific discovery.
Application Open
Ready to Collaborate?
Use the delivery form below to request TRE accounts for your institution. Describe your research proposal and how you intend to use the data assets - we'll review and contact selected teams during March-April 2026.
Pheno.AI & OPEN INNOVATION
Most comprehensive Deep-Phenotype Global Datasets
The Human Phenotype Project is a global, longitudinal deep-phenotype cohort. It integrates multi-omics, clinical, sensor and lifestyle datasets across large human cohorts to discover biomarkers, mechanisms and prediction models for human health.
Who is Pheno.AI?
Deep human cohort & knowledgebase
Pheno.AI operates the Human Phenotype Project globally, HPP integrates multi-omics, imaging, continuous monitoring and rich lifestyle data to support high-impact research on metabolism, microbiome, cardiometabolic disease and aging
HPP ecosystem
Prospective, longitudinal cohorts
Repeated visits every 2 years and synchronous data collection, allow mapping of trajectories and early signals before overt disease.
Collaborations
Collaborative culture
The Open Innovation program enables leading scientific teams to access the HPP datasets and conduct ground-breaking research.
Research proposals are selected based on their merit and teams can access the datasets through a trusted research environment
PLATFORM
AI-Native Trusted Research Environment
Projects are conducted inside a Trusted Research Environment with pre-configured compute, dataset loaders, notebooks and tools tuned for multi-omics, microbiome and time-series modeling.
Compute
Ready-to-use analysis environments
Access Python and R environments with Jupyter support, pre-configured for data science, ML and deep learning workflows.
Developer experience
Dataset loaders & notebooks
Dataset-specific loaders and example notebooks help teams quickly explore and work with available datasets.
Explainability
Model interpretability tools
Support for techniques such as Shapley-based feature attribution across clinical, molecular and behavioral features.
RESEARCH THEMES
Example Research Directions
Proposals can focus on these areas or introduce new ideas that leverage the multi-modal nature of the cohorts.
Metabolic health
Glycemic variability and lifestyle patterns
-
Characterizing glycemic variability and transitions using CGM, labs and lifestyle data.
-
Identifying early predictors of metabolic disease trajectories and related outcomes.
-
Evaluating longitudinal lifestyle patterns associated with resilient metabolic profiles.
AI & digital biomarkers
Multimodal predictive models
-
Developing models that combine multi-omics, imaging and time-series data into predictive scores.
-
Investigating explainable AI methods suitable for clinical decision support and trials.
-
Defining digital biomarkers that can be validated and replicated across cohorts.
Microbiome
Host–microbiome mechanisms
-
Associating microbiome signatures with metabolic, immune or inflammatory phenotypes.
-
Integrating microbiome with host genomics, metabolomics and diet for mechanistic hypotheses.
-
Exploring microbiome-related biomarkers with translational relevance.
Longitudinal health trajectories
Mapping the health–disease continuum
-
Studying transitions linked to aging, menopause or exposure to interventions.
-
Using continuous sensors to map trajectories across repeated visits.
-
Designing follow-up studies based on observed patterns and risk segments.
Open Innovation Starter Kit - Pheno.AI​
Explore Pheno.AI knowledgebase
Technical Knowledge base
The Pheno.AI Knowledgebase is the authoritative gateway to the Human Phenotype Project’s multimodal dataset ecosystem. It offers structured documentation across all data types - genetics, transcriptomics, metabolomics, microbiome, imaging, CGM, sleep, lifestyle, clinical measurements and more - along with data models, QC descriptions and analysis notes. Through guides, FAQs, and the pheno-utils library, it enables fast onboarding and reproducible research, making it the essential resource for navigating and working with HPP data.
​
Explore the full dataset documentation in the Pheno.AI Knowledgebase
Explore HPP Articles
Deep Phenotyping in the Human Phenotype Project / Nature Medicine (2025)
A landmark overview of the Human Phenotype Project (HPP), a 27,000-participant, deeply phenotyped cohort integrating genetics, transcriptomics, microbiome, imaging, CGM, sleep and lifestyle data over 25 years. Early findings highlight hidden glycemic variability, biological aging clocks, lifestyle-driven metabolic risk and high-accuracy disease signatures. AI foundation models such as GluFormer and COMPRER demonstrate HPP’s potential to transform precision health through multimodal prediction.
[🔗 Read the full article]