Network Bio and Duke Partner to Build AI Models of Human Disease
Biotechnology company Network Bio has announced a research collaboration with Duke University to develop artificial intelligence (AI) models that could improve disease detection, support drug discovery and help identify more effective treatments.
Under the partnership, Network Bio will work with Duke researchers to study tissue and blood samples stored in the university’s biobanks. The company will generate detailed molecular data from these samples and combine the findings with de-identified clinical records, including information about patients’ diagnoses, treatments and health outcomes.
The combined datasets will be used to train AI models that can better understand how diseases develop and respond to treatment. The research aims to support earlier disease detection, identify new drug targets and help researchers determine which treatments may work best for different groups of patients.
One of the major challenges in applying AI to healthcare is the availability of reliable human biological data. Many existing AI models in biology are trained using cell lines, animal studies or publicly available datasets. While these sources are useful, they may not fully reflect how diseases behave in real patients.
By combining biological samples with clinical information collected over time, Network Bio and Duke aim to give AI models a more complete picture of human disease.
The collaboration will focus on selecting disease-specific groups from Duke’s biobanks. Network Bio will carry out molecular profiling and sequencing of the selected tissue and blood samples. The resulting data will also be shared with Duke to support further academic research.
The partnership brings together Duke University’s biobanking resources and clinical research experience with Network Bio’s AI and data analysis capabilities. The two organizations also plan to establish a framework for using the datasets in research and commercial applications, including AI model development and data licensing, while following applicable privacy and data security requirements.
Asad Ali Ahmad, Co-Founder and CEO of Network Bio, said the collaboration would expand the company’s network of academic medical center partners. Its existing partners include the University of Pennsylvania, Mass General Brigham and the University of Colorado Anschutz.
Svati H. Shah, Associate Dean for Translational Research at Duke University School of Medicine and Director of the Duke Center for Precision Health, highlighted the potential of combining advanced molecular profiling, clinical information and AI to better understand diseases and how patients respond to treatment.
She added that the partnership could create new research opportunities while helping scientists make better use of existing biological samples and clinical information.
The collaboration also reflects the growing role of AI in pharmaceutical research and drug development. Combining molecular information with real-world patient outcomes could help researchers identify biomarkers, understand differences between patients and discover new therapeutic targets before potential treatments enter clinical trials.
For pharmaceutical and diagnostic companies, these datasets could provide valuable insights into disease biology, patient selection and the development of more targeted therapies.
Network Bio is based in Palo Alto, California, and is developing AI models designed to understand human biology using tissue, blood and clinical data. Its work focuses on applications in precision diagnostics, biomarker discovery and drug development.
Through its collaboration with Duke, the company aims to strengthen the use of human biological data in AI-driven medical research and support the development of more personalised approaches to healthcare.