NuRelm Helps Mine Cancer Data for Pitt Researchers

Mining a Mountain of Cancer Data for the Gem of Truth That Can Save a Life

NuRelm is honored to have recently worked on one of the first projects taken on by the Pittsburgh Health Data Alliance, a collaboration between UPMC Enterprises, Carnegie Mellon University’s Center for Machine Learning and Health, and University of Pittsburgh’s Center for Commercial Applications of Healthcare Data.

The project is called Tumor-Specific Driver Identification (TDI). This project began as an algorithm developed by researchers in the University of Pittsburgh’s Department of Biomedical Informatics. TDI is in a class of software that will provide personalized genomic information to cancer clinicians about the genetic drivers of an individual patient’s tumors. Tumor-specific algorithms will be used for real-time mining of genetic “big data” that will enable personalized treatments for cancer patients. TDI also is expected to lead to the discovery of new cancer drivers and may be used by pharmaceutical companies to identify novel drugs. Researchers on this project are Xinghua Lu, M.D., Ph.D., and Gregory Cooper, M.D., Ph.D., of the Department of Biomedical Informatics.

TDI Sample ScreenNuRelm helped the team take TDI from the lab to potential clients and collaborators. Our role was to develop technology around a user interface, API (application programming interface), result storage / management, reporting,  and systems to scale up to allow multiple intensive TDI runs to happen simultaneously in a production environment. The result is a fully functional, auto-scaling, self-healing system hosted on Google Cloud Platform.

Sam Shaaban, NuRelm’s co-founder and CEO, recently presented a tech talk on this project at the annual Big Data conference in Pittsburgh.


  1. Congratulations NuRelm! That is quite an accomplishment. This has the potential to help so many and is much needed today. MAny people will thank you, now and in the future!!!!!
    Jim and Caroline Stuncard

Leave a Reply

Your email address will not be published. Required fields are marked *