Skip to main content
Subha Madhavan


Subha Madhavan, PhD

Current Affiliation

Chief Data Scientist, Georgetown University Medical Center, Washington, D.C.


BS, Birla Institute of Technology and Science
MSc, Birla Institute of Technology and Science
PhD, Uniformed Services University for the Health Sciences (Indo-US Collaborative Graduate Program)
Postdoctoral Fellow, Johns Hopkins University School of Medicine, Baltimore
MS, Computer Sciences and Information Systems, University of Maryland

How do I describe my work to those outside the field

What resonates with people who are not informaticians is that I work at the interface of computer science, biology, and medicine. My team and I collect a lot of data, we analyze and visualize information with the goal of extracting actionable knowledge to improve our understanding of diseases, of research overall, how we provide care today, and to improve patient outcomes efficiently. That’s what we do! I don’t like to use big buzz words, like ‘machine learning’ or ‘artificial intelligence’. This conveys the message to people and they go ‘ah ha! I get what you do’. I think it’s so important to convey what we do in lay person’s terms, because decision makers, such as sponsors, Deans, are not usually informaticians. It is important for us to be able to communicate using simple but powerful words that align with the organizations goals.

Years of experience:

I started in informatics in 2003.

Why Informatics?

I was a post-doctoral fellow at Johns Hopkins University after my PhD in Molecular Biology and I was running large radioactive protein gels to study proteins involved in anemia and other blood-related disorders. That was the primary focus of our laboratory. I was going through a millicurie of radioactivity every week, which is a lot, and I was exploring how to contribute to medicine beyond being a lab rat. I had an undergraduate degree in Chemical Engineering with statistics and programming experience. This allowed me to collaborate with colleagues at Hopkins to analyze gene sequences, which was a hot field at that time, 1999-2000, as the human genome was being sequenced around the world, and there was a lot of energy and excitement around this. A lot of people were beginning to look at these genome sequences and translated protein sequences to understand disease and they needed help analyzing and interpreting these sequences. I really enjoyed being the ‘informatics’ collaborator with my colleagues at Hopkins. That experience really launched me into informatics and data science which led me to go back to school at the University of Maryland and get a degree in Computer Science so that I could be formally trained in the computational field.

That Hopkins experience really made me realize that I have a passion and interest for this, but I had to learn additional skills, so after this I started working for a small business startup, which for the very first time was launching software products in bioinformatics to allow users to conduct pathway analysis and gene expression analysis and interpretations. I moved from there to the National Cancer Institute where I could apply the knowledge of software and technology and informatics process to building clinical databases like Rembrandt, which was one of the first clinical-genomic brain tumor databases. We collaborated with clinicians and software engineers and data curators for this project and it was an exemplar nationally-run informatics program. The program won the Service to America award. This project could not have happened without informatics. I became a card-carrying informatician after the Rembrandt project as I helped to match the clinical data with the genomic data from patients with brain tumors, made the data set available publicly for researchers, for analysis by our own team and the larger biomedical community. This is the power of informatics. You can connect the dots, you can bring people together, and we can extract and utilize actionable knowledge in discovery, translation and care. I’m so lucky to have the opportunity to do this every day in my life.

What are your ambitions? At the end of your career, what do you hope to have accomplished?

My ambition is to have meaningful impact in clinical care. This could be in improving research that leads to better treatments, or implementation science that leads to better clinical pathways or policy that impacts public health. To be able to translate our methodologies and our data applications into real world use. We are moving the needle ever so little every day, but impacting clinical guidelines and health policy is not easy, and this is where I think informatics has a big role to play because we want to bring all the evidence from the data to influence these communities of clinicians, administrators and policy makers. I would be very happy if our work translates directly to improve patient outcomes and deliver healthcare in an economically feasible and ethical fashion. That I would call a success.

Who or what are your “key sources” in the informatics field?

I follow a lot of luminaries in the field. Twitter is a great medium. Every morning I browse through my Twitter feed for experts in the field that I follow, and that gives me a daily digest of what information has been published, what our experts are saying, active debates that are going on and conferences that are trending, all in 140-character-sound bites allowing me to drill down as needed. AMIA, of course, has been a great source of collaborators, influences and role models for me over the years. I do not miss a single conference of AMIA, both the Summit and the main Symposium. I had the great honor of being the scientific program chair for the AMIA Informatics Summit a few years ago and worked closely with the scientific planning committee to put that meeting together, so I really understand the effort that goes into curating a great conference, and I love to take advantage of that twice a year, to really learn what’s going on in the field and how we might apply that to our work. We want to think globally but act locally and these conferences have been tremendous sources of that information. And then I often go to journal clubs, grand rounds, think tanks and seminar presentations not just in informatics, but also in other disciplines, such as medicine, public health, cancer research, etc., because we need to understand how our work can impact and influence and support other domains. Our own discussions within the Informatics Center are important and I think it is equally important to continue to learn from other domains so that we can really apply the work that we are doing, and also inform their work.

Articles that spotlight my research interest

  1. Adapting crowdsourced clinical cancer curation in CIViC to the ClinGen minimum variant level data community-driven standards. Danos AM, Ritter DI, Wagner AH, Krysiak K, Sonkin D, Micheel C, McCoy M, Rao S, Raca G, Boca SM, Roy A, Barnell EK, McMichael JF, Kiwala S, Coffman AC, Kujan L, Kulkarni S, Griffith M, Madhavan S, Griffith OL. Human mutation. 2018; 39(11):1721-1732. PMID:30311370; PMCID:PMC6282863
  2. Molecular Profiling of Patients with Pancreatic Cancer: Initial Results from the Know Your Tumor Initiative. Pishvaian MJ, Bender RJ, Halverson D, Rahib L, Hendifar AE, Mikhail S, Chung V, Picozzi VJ, Sohal D, Blais EM, Mason K, Lyons EE, Matrisian LM, Brody JR, Madhavan S, Petricoin EF 3rd. Clinical cancer research : an official journal of the American Association for Cancer Research. 2018; 24(20):5018-5027. PMID: 29954777
  3. The REMBRANDT study, a large collection of genomic data from brain cancer patients. Gusev Y, Bhuvaneshwar K, Song L, Zenklusen JC, Fine H, Madhavan S. Scientific data. 2018; 5:180158. PMID:30106394; PMCID:PMC6091243
  4. viGEN: An Open Source Pipeline for the Detection and Quantification of Viral RNA in Human Tumors. Bhuvaneshwar K, Song L, Madhavan S, Gusev Y. Frontiers in microbiology. 2018; 9:1172. PMID:29922260; PMCID:PMC5996193
  5. Art and Challenges of Precision Medicine: Interpreting and Integrating Genomic Data Into Clinical Practice. Madhavan S, Subramaniam S, Brown TD, Chen JL.American Society of Clinical Oncology educational book. American Society of Clinical Oncology. Annual Meeting. 2018; (38):546-553.PMID:30231369

Hobbies/Interests outside AMIA

I love to Zumba! I used to do classical Indian dance but I enjoy dancing in my Zumba exercise classes. It’s like free-flowing dance for the body. I love to cook, especially fusion cuisines.

AMIA is important to me because

AMIA is a place where I go to problem solve. For me, getting input from other leaders in the field, and from other colleagues who are probably facing similar issues, both from an administrative standpoint, or career development standpoint, or a scientific problem that I’m trying to solve. I tend to try to meet with people who I cite in my papers or with potential collaborators. I also participate in a number of AMIA working groups, which are these focus, special interest groups, genomics and translational bioinformatics, or clinical research informatics, or the academic leaders in informatics. It really provides a venue for communities to come together and problem solve. That’s what AMIA means to me.

I am involved with AMIA

I chaired the Informatics Summit Scientific Program Committee a few years ago. I’m a member of the Genomics and Translational Bioinformatics working group, and the Academic Leaders in Informatics group.

It may surprise people to know

I came to the U.S. with $300 in my pocket as an immigrant. This is who I am. Twenty-two years ago I came to the United States from India as a grad student. I come from humble beginnings in India and still feel that parts of me are still that resource-constrained graduate student. That influences how I run my Center and advise industry partners. Knowing how to do more with less is especially important now. We need to be more creative than ever in finding the right resources for data science.