Early-stage research exploration
An emerging collaboration between biomedical researchers and AI systems engineering, exploring longitudinal health data, early disease signals, and computational biology.
Meet the research collaboratorsBiological signals rarely arrive in one place. They emerge across molecular measurements, observations over time, and a growing body of scientific literature.
We are exploring how an evidence-connected intelligence system could help researchers connect these layers, frame testable hypotheses, and investigate the biological context behind a signal.
The work is at the stage of defining research questions and possible collaborations. No clinical performance or validated disease-detection capability is claimed.
Biological interpretation, genomic context, and the engineering to bring evidence together.
Hematology & translational biology
Professor · Kathryn Mays Johnson Distinguished Chair in Oncology, UT Health San Antonio
Studies blood cancers, mitochondrial metabolism, and tumor immunology, with an emphasis on translating biological mechanisms into therapeutic approaches.
His work includes MDS disease models and cell-based approaches to solid tumors. He also founded Powerhouse Therapeutics to advance laboratory inventions toward application.
Genomics & epigenetics
Assistant Professor · Cell Systems & Anatomy, UT Health San Antonio
Investigates how genetic variation, epigenetic regulation, and 3D genome organization influence cell identity, disease, and therapeutic response.
Combines functional genomics with computational modeling. Recognition includes the 2024 CPRIT Scholar and UT System Rising STARs awards.
AI systems engineering
Founder & CEO · Flow Mind AI Inc. / Principal Software Engineer · Palo Alto Networks
AI systems builder with 11+ years of experience in enterprise backend, platform, and LLM-powered systems. A Principal Software Engineer at Palo Alto Networks and founder of Flow Mind, the company behind Mad Dog Alpha.
Takes AI systems from prototype to production, combining distributed systems, API design, and cloud-native infrastructure with RAG, context engineering, and LLM orchestration. Builds evaluation pipelines, regression checks, guardrails, and observability to make AI workflows reliable and maintainable at enterprise scale.
Explore how longitudinal observations could reveal patterns that isolated measurements miss.
Connect candidate signals with gene regulation, cell states, and disease mechanisms.
Keep source evidence and uncertainty visible as researchers develop and evaluate hypotheses.
Prior academic work and awards of individual collaborators, providing context for this exploration.
Nature / 2024Yu Luan CPRIT Scholar Award · UT System Rising STARs Award, 2024
View faculty recordBiomedical Research Initiative
For research collaboration, early-stage investment conversations, or an introduction to the initiative, get in touch with Deheng.
Discuss the initiativeEarly-stage research exploration. Institutional affiliations do not imply institutional endorsement.