TESA Research uses advanced machine learning to analyze the full genomic architecture of cancer, not just individual variants, enabling smarter trials and identifying patient populations where deprioritized drugs may still succeed.
Tumors are heterogeneous. When promising therapies are tested in broad unselected populations, real responses from the right patients get diluted away, and effective drugs fail.
We are not failing because drugs don't work. We are failing because we don't know who they work for.
Our proprietary platform uses neural-network models to integrate complex multi-omic data. We look at structural genomics, transcriptomics, sequence variants, copy-number variation, and epigenetics to identify recurrent biological patterns that correlate with disease course and therapy response.
Most biomarkers read one variant. Tumor behavior emerges from interactions across the genome. That's what we model.
Statistically significant prognostic and predictive signatures demonstrated across five cancer types with leading academic collaborators.
Built by a team that has shipped genomic software used in clinical labs worldwide. We understand both the science and the implementation path.
Enrich ongoing trials, revisit halted assets, or explore patient cohorts with a platform-agnostic computational partner.
Explore partnerships →A proven, exited founding team; a validated platform; and a business model built around the industry's largest untapped resource.
Request investor materials →Work at the frontier of AI and cancer genomics with a small founding team that has built and exited before.
See open roles →Whether you're validating drug targets, exploring patient cohorts, evaluating an investment, or looking for your next role, we'd like to hear from you.