UNC biostatistics study deepens understanding of genomic puzzle

June 01, 2012
Dr. Matthew Nelson

Dr. Matthew Nelson

Researchers at UNC Gillings School of Global Public Health have helped usher in a new understanding of genetic variants and disease by conducting one of the largest population studies of genome sequencing to date.

Matthew R. Nelson, PhD, adjunct associate professor of biostatistics and a principal scientific investigator of statistical genetics at GlaxoSmithKline, is lead author of “An abundance of rare functional variants in 202 drug target genes sequenced in 14,002 people,” published in the May 17 online edition of Science. In the work, Nelson and others from UNC and the University of California at Los Angeles describe how “rare” variants in the human genome are actually quite common, and may contribute to diseases like diabetes, bipolar disorder, cancer and Alzheimer’s.
“There’s an overabundance of rare genetic variation,” Nelson said. “And we demonstrate in this paper that much of this variation is probably damaging.”
In other words, Nelson said, rare variations in our genomes aren’t so rare after all, and they are probably to blame for many of the diseases that plague us. However, because of rapid human population growth in the last 10,000 years, natural selection has not had an opportunity to remove these variants from the genome.
Researchers studied 202 genes (potential targets for drug therapies) found in 14,002 people, which means they studied 864,000 base pairs out of the three billion present in the human genome. That’s a small slice of the genome pie, but it gave Nelson and colleagues significant insight into how common rare variants are. Researchers found one rare genetic variant in every 17 base pairs, a startlingly frequent occurrence.
Nelson’s paper marks a change in thinking about the human genome. When scientists first began sequencing our DNA in the hopes of finding variants that caused disease, they expected to find a common number of changes and deviations that could be identified easily and targeted with drugs. Now, Nelson’s work shows the magnitude of difference contained within our DNA.
“Now we can learn more about the role of those genes in disease,” he said.
Nelson said much of the study’s early work, which began in 2009, was completed with the help of paper co-authors Zhengzheng Tang, a UNC biostatistics doctoral student, and Yun Li, PhD, an assistant professor of genetics and biostatistics at UNC.
“A lot of key work was determining how reliable these data were,” Nelson said. “When we started to find all the rare variants, we were concerned to see if variants were real.”
Michael Kosorok, PhD, professor and chair of the biostatistics department and professor in the department of statistics and operations research, said Nelson’s study is truly multidisciplinary and highlights the important role biostatistical analysis can play in understanding genetics.
“Biostatistics seeks to find tools to analyze data and tools for establishing the degree of certainty of the results” Kosorok said. “It’s a very important study that’s already received a lot of national attention.”


UNC Gillings School of Global Public Health contact: Linda Kastleman, communications editor, (919) 966-8317 or linda_kastleman@unc.edu.