As most scientists know, small amounts of tissue can hold a treasure trove of molecular and genetic information that can help them better understand disease processes and how to stop them.
Today, the science of bioinformatics is harnessing powerful information technology tools and computational techniques to extract meaning from large volumes of raw research data, providing huge amounts of patient data information that will help inform more precisely tailored treatments.
“The bigger the data, the more precise are the research and clinical answers,” said Reddy Gali, a bioinformatics educator on Harvard Medical School’s new C3 bioinformatics team.
But such large amounts of data can be overwhelming, and computational expertise is necessary to get the researcher’s questions answered. The tools and computing power of bioinformatics are so essential to modern research that it supports more than 50 percent of every biological research study, Gali said.
C3 stands for Harvard Catalyst, Countway Library and Center for Biomedical Informatics (recently changed to the Department of Biomedical Informatics), a joint venture that is helping Harvard’s scientists and affiliated hospitals meet the increasing demand for bioinformatics training and consultation services.
The field of bioinformatics has become so integral to the mission of reducing the burden of human disease that Jeffrey S. Flier, dean of Harvard Medical School, recently transformed the Center for Biomedical Informatics into a fully-fledged academic department.
“Everywhere you go nowadays, you need bioinformatics to support basic and translational research,” said Gali, who teaches most of the workshops that C3 Bioinformatics conducts at Countway Library every year.
Large labs may have their own bioinformatics experts, but for most labs, the need is not constant.
“That’s why Harvard Catalyst is supporting this program—one group of us is available to help hundreds of researchers every year,” Gali said.
Tools and Workshops
Hundreds of bioinformatics tools are available through C3. The most popular, according to interest in their workshops, Gali said, is next-generation sequencing analysis tools through which gene variants, DNA-protein binding regions and the expression of genes are analyzed.
Typically the C3 team, which also includes Countway librarians David Osterbur and Paul Bain, runs about 70 hands-on workshops every year.
Gali said he is also available to the Harvard community to consult on applying bioinformatics to every phase of the research process, from designing an experiment and acquiring data to analyzing and interpreting results.
“Without Reddy’s help, I don’t think I could have found the molecular pathway and target genes to follow as I did,” said Yu-Hwa Huang, an HMS instructor in medicine at Brigham and Women’s Hospital.
In a major paper published in Nature earlier this year, Huang reported the identification of two interacting molecules that inhibit T-cell function. By blocking this pathway, she and her colleagues were able to boost immune response and inhibit colorectal cancer progression in an animal model.
That paper reported on the nanostring analysis that Huang performed to detect expression of 200 genes related to immune response in this newly identified pathway.
Previously, she had done a microarray analysis to see what genes were influenced when the pathway was blocked. All the data these analyses generated filled a CD, however, and Huang wasn’t sure how to proceed. She attended one of Gali’s workshops and met with him several times over the next three years.
“He’s my hero in bioinformatics,” she said.
Yiguo Hu, a research fellow in medicine at Dana Farber Cancer Institute, learned to use microarray to compare normal versus tumor tissue. He then developed a mouse model to better understand and test therapies to stop the disease process of multiple myeloma, a cancer that forms in white blood cells.
Hu felt that Gali contributed so much to the analysis of his research results that Hu named Gail a co-author on a paper Hu published recently.
“Now, next time I can do the data analysis on my own,” Hu pointed out.
Though most researchers come to Gali after they’ve already generated data, like Huang and Hu, Gali recommends that researchers consult with a bioinformatics expert during the design phase of their experiments.
That’s what Mary Frances Lopez, HMS assistant professor of pediatrics at Boston Children’s Hospital, did. Lopez didn’t have any background in bioinformatics and decided to attend a microarray workshop before she designed her study. She then worked with Gali, who guided her one step at a time.
“We are identifying genes that are important for growth and found some metabolic pathways that the mice were lacking,” Lopez said.
“Without bioinformatics, we wouldn’t be able to identify the genes and explore how these genes affect fetal growth, obesity and cancer,” she said.
Often, for clinical research, large numbers of study participants are needed for statistical power. Harvard Catalyst sponsored the creation of a web-based query tool to determine all patients who meet a given set of inclusion and exclusion criteria from five of the largest Harvard-affiliated medical centers—Massachusetts General Hospital, Brigham and Women’s, Boston Children’s, Dana Farber and Beth Israel Deaconess Medical Center.
Called SHRINE (Shared Health Research Information Network), it enables researchers to search for privacy-protected patient demographics, diagnoses, medications and laboratory results.
The data generated are useful for investigators who are interested in creating new research hypotheses, Gali pointed out, adding that he offers a SHRINE workshop every month.
Bioinformatics has expanded well beyond its initial application in genetics.
“The amount of data is exploding,” commented Gali, “and along with it, the demand for bioinformatics.”