Addressing the Data Bottleneck

Collaboration: Addressing the Data Bottleneck

By Dan Fost

In an age of tremendous biological discoveries, where stem cells may ultimately prove useful as therapies for many diseases, a quiet mathematician and physicist is emerging as a key player.

Jun Song, PhD, has found himself in demand among the biologists in the Eli and Edythe Broad Center of Regeneration Medicine and Stem Cell Research at UCSF, who value his ability to process and interpret the massive data sets generated by their research.

Song was on the verge of moving his office across San Francisco, to the recently established UCSF Mission Bay campus on the city’s eastern edge, when the stem cell researchers found room for him in their new building at Parnassus Heights.

“He had already started several collaborations here, and if he moved, this would have meant the loss of his expertise,” says Arnold Kriegstein, MD, PhD, director of the stem cell center. Keeping him, Kriegstein adds, will maintain existing partnerships and likely lead to new ones.

One of Song’s most fruitful collaborations started with a serendipitous car ride in April 2010.

Miguel Ramalho-Santos, PhD, MSchad invited Song, an assistant professor in UCSF’s Institute for Human Genetics, to speak at a stem cell conference in Monterey.

As Song drove Ramalho-Santos to the conference in his Honda Fit, the scientists spent the two hours talking about conducting a sophisticated comparison of human induced pluripotent stem (iPS) cells – those derived from adult cells – and human embryonic stem cells.

Since both types of cells can grow into all cell types, Ramalho-Santos said that knowing how they differed would be a critical underpinning of future work. What they found, Ramalho-Santos says, was that the adult cell remembers its roots. “There are traces we can find of gene activity in an iPS cell that can tell us where it came from,” he says.

“The bottleneck in research could become data analysis,” Song says. “The amount of data will dramatically increase in the future.”

Even though computational power is also growing, integrating and interpreting the data remain difficult problems. What is needed are scientists who understand quantitative data as well as cell biology. Song thinks it will be important to teach biology to computational science whizzes, in addition to training biologists to think quantitatively.

“I want to help people who are like me in their career path,” Song says, “and want to venture into a new field and are interested in solving biological problems."