One of the most unexpected and dangerous side effects of drugs is Long QT Syndrome, a change in the electrical activity of the heart visible on the electrocardiogram that increases the risk of a potentially fatal arrhythmia called torsades de pointes.

The effects of individual drugs on the QT interval are well-studied, but until recently little was known about how multiple drugs taken at the same time could interact to cause this side effect. As part of my PhD I developed methods to computationally predict and experimentally validate drug-drug interactions leading to Long QT Syndrome.

To allow others to make their own discoveries leveraging the same big data we used, I worked with a team to create ∆QTDb, an explorable database of the effects of drugs on the QT interval.

To create this resource I first analyzed and processed anonymized drug exposure and electrocardiogram data from our electronic health record. The resulting MySQL database contains each given patient's change in QT interval after exposure to one or more commonly prescribed drugs. I then implemented a web app using React.js and d3.js and an API to retrieve and visualize the relevant subset of data from the database.

We're currently preparing ∆QTDb for publication; in the meantime check out the site at www.deltaqt.org!

∆QTDb sketch
∆QTDb screenshot

Team: Tal Lorberbaum, Victor Nwankwo, Nicholas Tatonetti. I conceived of ∆QTDb with Nick; created the database; and implemented the web app with Victor.