Artificial intelligence can feel like a field that’s impenetrable and huge — it can be hard to know where to get started. That’s why we’ll be introducing you regularly to someone working in AI who will share their story of getting involved in AI. This month, we had the pleasure to interview Dr. Marina Sirota, an assistant professor at the Institute for Computational Health Sciences at UCSF, where she discovers new diagnostic strategies for disease using computational methods.
AI4ALL: Can you tell me a little bit about what you’re working on right now? What problems are you most interested in solving?
MS: Given the wealth and availability of -omics data, computational integrative methods provide a powerful opportunity to improve human health by refining the current knowledge about the determinants of disease. My research interests lie in developing computational methods integrating a diverse set of molecular and clinical data and applying them to discover novel diagnostic and therapeutic strategies. More specifically, I have a strong interest in studying the role of the immune system in health and disease. Can we come up with better and faster ways to diagnose autoimmune disease? Can we use genomic data to identify novel therapeutic targets or new uses for existing drugs?
How did you get interested in this work? And how did you get interested in AI?
I actually got interested in the field of bioinformatics when I was in high school. I was pretty good in math and science and thought that I would likely go into software engineering. As a junior in high school however I took a biotech class where we got to do cool experiments such as transfecting Green Flourescent Protein (GFP) into bacteria making it glow in the dark. I was fascinated with the field of genetics and wanted to bridge my interest in biology and computer science when I started my undergraduate at Stanford.
Proudest or most exciting moment in your research?
The most exciting moment in my own research was when I was a graduate student and worked on developing a systematic computational approach to predict novel therapeutic indications for existing drugs leveraging genomic data. We applied our computational approach to discover new drug therapies for inflammatory bowel disease (IBD) and collaborated with an experimental group to test our predictions. Among the top compounds predicted to be therapeutic for IBD by our approach were prednisolone, a corticosteroid used to treat IBD, and topiramate, an anticonvulsant drug not previously described to have efficacy for IBD or any related disorders of inflammation or the gastrointestinal tract. Using a rodent model of IBD, our collaborators experimentally validated the efficacy of topiramate in vivo. This was the first time a computational prediction I made was validated in the lab, which was tremendously exciting! This approach has now been applied to numerous disease indications including lung and liver cancer, dermatomyocitis and has inspired several startups.
The proudest moment so far for me as a researcher and mentor, however, was when a high school student that I mentored last summer, Anooshree Sengupta, presented her work at the AMIA conference this past October.
She was one of 5 high school students who were chosen to present their work at the conference and did a fantastic job!
Who were your role models growing up?
I know this is cliche, but my parents were my role models growing up — my dad is an engineer and a researcher and my mom is an educator in the area of computer animation and visual effects. They always encouraged me to follow my passion which led me to where I am today.
Where do you see AI making the biggest impact in the next 5 years?
AI has already had a huge impact and transformed multiple industries — Google has transformed ways that we obtain information, Coursera has enabled free online learning, LinkedIn and Facebook have transformed the way we communicate with each other, Uber and Kayak have made transportation and travel easy and accessible.
I truly believe that AI will have a huge impact on healthcare in the next 5 years — from research areas leveraging -omics data for diagnostics and drug discovery to more applied areas of leveraging electronic medical record data to aid clinicians, this is a huge opportunity for the application of AI and I’m very excited to be part of this journey!
What advice do you have for young people who are interested in AI?
I would advise young people who are interested in AI to follow their passion and learn more. Don’t get discouraged or overwhelmed by the complexity of the methods. Most important, don’t be afraid to ask questions — that’s how you will learn! Try to find an exciting problem and work on it. Find something that really pisses you off and try to solve it with the technology that is exciting to you — this is what will drive true innovation and make an impact.
Marina Sirota is an assistant professor in the Institute for Computational Health Sciences at UCSF in San Francisco, CA, and has worked as a research scientist at Stanford University and Pfizer. Her primary focus is on leveraging and integrating different types of -omics and clinical data to better understand the role of the immune system in disease. See what her lab is up to here.