Researchers Learn to Harness Data to Produce Better Science

A new educational program aims to teach researchers ways to handle a large amount of data.

Scientists digging into large amounts of data need the right tools at the right time to uncover helpful nuggets that could yield creative insights to produce better science.

Children’s Hospital of Philadelphia and Drexel University are collaborating to help educate all researchers, not only those at CHOP, to fill their virtual toolboxes.

The team will create a Biomedical Data Science education program, with the help of a new National Institutes of Health grant, that provides evidenced-based pathways and learning modules to guide researchers who are working with data-intensive science. Their goal is for the program to be as broadly applicable and usable by as many people as possible so that they are well prepared to use NIH funding, can be better stewards of knowledge and science, and can pursue pediatric research that is more reproducible and reflective of scientific rigor.

“CHOP recognizes that for pediatric research to really accelerate, we need to be a leader in how we use data and how our researchers use data,” said Jeff Pennington, associate vice president and chief research informatics officer at CHOP Research Institute, who is the principal investigator on the grant. “That recognition led to some major strategic investments in our data science programs. The pediatric angle is especially important because although there is a lot of investment by pharma and the NIH for adult health conditions, the funding for pediatric health conditions is much lower. We’ve had to be much better to do more with what we have, and to invest in data science to make more progress toward improving child health.”

The program will be a modular style that allows researchers to mix and match the courses based on what makes sense for their research and the type of data that they’re working with. It will consist of a combination of modalities, such as screencasts, code snippets, articles, and recorded seminars, that will accommodate a variety of learning types with a wide range of cognitive and physical abilities.

“What a principal investigator needs to learn to help write a research data management lifecycle plan is different from what a study coordinator needs to learn to track recruiting,” said Joy Payton, supervisor of data education on the Arcus team. “The program will allow users to construct a learning structure that makes sense for them, and to learn something that is useful to put into practice right away.”

The cross-disciplinary project will build on an existing partnership between the Arcus team in the Department of Biomedical and Health Informatics at CHOP and College of Computing and Informatics at Drexel University. Ali Shokoufandeh, PhD, professor of computer science at Drexel, said that teaching data science to professionals is different from teaching undergraduate and graduate data science students, and that is one exciting aspect of this new program.

“This challenges us at Drexel as an academic institution to adapt our program so that the knowledge can be transferred and used in a professional setting in an as-needed education or training program,” Dr. Shokoufandeh said. “One of our objectives is giving people who don’t have data or computational literacy the skills to handle massive amounts of data. If you’re doing any sort of research or science, and you don’t have the tools to do this, you’re at a disadvantage. But, you don’t need to be a computer scientist or data scientist to be able to handle your data.”