Informatics×

News

Phyllis Thangaraj
Phyllis Thangaraj, MD/PhD student (Tatonetti lab)

Aspiring physician-scientists from Columbia's Vagelos College of Physicians and Surgeons presented their research posters at the 14th annual MD-PhD Student Research Symposium on April 25. Their research delved into a range of topics, including Alzheimer’s disease, stroke, and stem cells. The event included a guest lecture by an alumna about her own career path as a physician-scientist, and culminated in the poster session judged by MD-PhD alumni who currently work at the University. Department of Systems Biology’s Phyllis Thangaraj, an MD/PhD student in the Nicholas Tatonetti lab , was named one of five poster winners at the event. 

She presented work on applying machine learning methods to phenotype acute ischemic stroke patients in the electronic health records. In cohort research studies, it is essential to identify a large number of subjects in an accurate and efficient manner, but often this requires time-consuming manual review of patient charts. 

“We applied machine learning methods to data within a patient’s electronic health records to develop a high-throughput way to define research cohorts,” explains Thangaraj. “Our test case is in acute ischemic stroke. We extracted clues within a person’s medical record that required minimal data processing to classify those who have had a stroke. In a separate cohort, the UK Biobank, we were able to use our model to identify patients with self-reported stroke but no mention in their medical data with 65-fold better precision than random selection of patients.” Although stroke was the test case in this particular work, she explained that their workflow could be applied to identify patients for cohorts of other diseases, particularly when the dataset has missing data. 

Nicholas P. Tatonetti, PhD, has recently been named director of clinical informatics at the Institute for Genomic Medicine (IGM) at Columbia University Medical Center. In this new role, he is charged with planning, organizing, directing and evaluating all clinical informatics efforts across the Institute. In particular, he will focus on the integration of electronic health record data for use in genetics and genomics studies.

Dr. Tatonetti, who is Herbert Irving Assistant Professor of Biomedical informatics with an interdisciplinary appointment in the Department of Systems Biology, specializes in advancing the application of data science in biology and health science. Researchers in his lab integrate their medical observations with systems and chemical biology models to not only explain drug effects, but also further understanding of basic biology and human disease. They focus also on integration of high throughput data capture technologies, such as next-generation genome and transcriptome sequencing, metabolomics, and proteomics, with the electronic medical record to study the complex interplay between genetics, environment, and disease.

At the Institute for Genomic Medicine, researchers are focused on innovative approaches to genomic medicine. Their multi-tiered approach to genomic medicine utilizes large scale genomic sequencing and analysis, paired with functional biology to advance the diagnosis, characterization, and treatment of genetic diseases. IGM is playing a critical role in Columbia’s overall Precision Medicine Initiative, a major University-wide effort to provide medical diagnosis, prevention and treatment based on an individual’s variation in genes, environment, and lifestyle. 

Dr. Tatonetti, who joined Columbia in 2012, is also affiliated with the Center for Computational Biology and Bioinformatics, the Department of Medicine, the Department of Biomedical Informatics, and the Center for Cancer Systems Therapeutics.