Startup Ozette unveils AI approach to understanding the immune system

Ozette applies artificial intelligence to immune monitoring to gain greater insight into a cancer patient’s immune system. Beyond informing treatment decisions, the technology can guide drug and diagnostic development.

AI, machine learning


Cancer patients receive a check-in to see how their immune cells are working and whether the disease is responding to treatment. This immune monitoring is a routine part of cancer care and it helps guide treatment decisions. Yet current analysis methods remain limited, according to Ali Ansary, co-founder and CEO of startup Ozette.

The Seattle-based company aims to take immune monitoring further and it is using artificial intelligence to do it. The approach could accelerate analysis of the immune system, helping clinicians make better diagnostic and treatment decisions. The technology could yield insights not only for cancer care, but also infectious disease, inflammatory disorders, and more, said Ansary, who is an attending physician at the University of Washington Medical Center.

Ozette is now unveiling its technology along with $6 million in seed funding. The financing was led by Madrona Venture Group, with participation from the Allen Institute for Artificial Intelligence (AI2) and Vulcan Capital. Ozette incubated within AI2.

Immune monitoring is currently done by taking a blood sample from a patient and looking at single cells to see whether a patient’s immune cells are functioning and how they’re responding to a therapy. Single-cell analysis isn’t new, but it is producing more data and a fraction of it is being used. Ozette ramps up the analysis by applying AI.

Ozette has developed algorithms to understand cell types, said Greg Finak, Ozette’s co-founder and chief technology officer. The company applies AI to sift through the data to provide a better understanding of what’s happening with cells, measuring things that weren’t previously possible. The technology also accelerates and automates the process. Work that would take months with current techniques can be done in hours, Finak said. Quick insights from single-cell data can then be used to inform patient care.

“What our technology allows us to do is to now see the cells of the immune system at several orders of magnitude higher,” Finak said. “We can get a bigger, better picture of what the immune system is doing than was previously possible.”

The Ozette technology comes from the Fred Hutchinson Cancer Research Institute in Seattle where Finak is a scientist. The company’s founders did occasional consulting work for clients looking to manage their data. Finak said those clients were looking for help extracting information from their data but as a research organization, Fred Hutch is limited in what it can do. Some clients preferred to engage with a company. The growing demand for data insight led the founders to form Ozette.

Ansary said that Ozette has built its technology with between $6 million and $8 million in grant funding. The seed financing is for building the company. The insights Ozette’s technology uncovers about immune cells could be used to develop new therapies and diagnostics, but Ansary said that’s not where his company is going—at least not at first. He describes Ozette as a technology company in the life sciences. The startup’s first focus is finding immuno-oncology insights that can improve the care of cancer patients. The near-term plan is to continue to develop the technology and establish partnerships with cancer research centers and companies.

Ozette is one of a growing number of companies using AI to gain insight into the immune system. Adaptive Biotechnologies, a fellow Seattle company with roots at Fred Hutch, is working with Microsoft to decode the immune system. Their research focuses specifically on the T-cell receptor. Last year, startup Immunai launched and unveiled its single-cell analysis technology that is the basis for its efforts to build a map of the immune system. The New York-company announced a $60 million Series A round of funding last week that it will use to discover and develop better, more targeted immunotherapies.

Ozette currently employs eight but Ansary said he expects headcount will double in the next few months, and double again by early next year. True to the interdisciplinary nature of the startup, those hires will come from a wide range of disciplines including data science, computational immunology, and software engineering.

“What we’re trying to do is build a really strong culture here that’s diverse in perspectives, diverse in backgrounds,” Ansary said. “That’s one of the core values we’re establishing here.”

Photo: ANDRZEJ WOJCICKI, Getty Images