Five fast facts you should know about Alzheimer’s research on the Blue Waters supercomputer

04.27.16 -

by Susan Szuch

Mariet Allen is an assistant professor of neuroscience at Mayo Clinic in Florida, and is conducting ongoing research using the Blue Waters supercomputer at the National Center for Supercomputing Applications (NCSA) at the University of Illinois to understand the role that genetics plays in Alzheimer’s disease. Here are five fast facts about her research.

1. This wouldn't have been possible without Blue Waters

Considering that the research is examining pairs of variants instead of individual variants, running these analyses are computationally intensive.

Because of that, Allen collaborated with investigators at the University of Illinois at Urbana-Champaign, receiving support from NCSA and the Mayo Clinic in order to run the analysis that was not possible without a supercomputer.

"It's not like the kind of analysis that can be done routinely, in most university institutions," Allen says. "That's where Blue Waters really came in and enabled us to do the analysis."

2. It's one of the only studies to analyze epistasis on this scale

The study looks at over 200,000 single nucleotide polymorphisms across whole genomes, in 350 different people. In addition to that, association of expression was tested on about 20,000 different genes. There is only one other report of a study of this scale, according to Allen.

"For each individual gene that we're looking at, we're performing more than a billion tests," Allen says. "It's more the scale of the analysis that requires Blue Waters."

3. It could help identify the genes that influence the most common type of Alzheimer's

Late-onset Alzheimer's is a complex disease, with multiple genetic and possibly environmental factors that are not yet completely understood, Allen says.

"Unfortunately, it's not one of those diseases where you say OK, this gene has a mutation and now they will certainly have Alzheimer's disease because this protein is has been altered in this person," Allen says.

By contrast early-onset Alzheimer's disease is more rare, present in around 5 percent of people with Alzheimer's. In some cases, it's known that this can be caused by a genetic mutation that is inherited from parents.

In late-onset Alzheimer's, the causes are believed to be more complicated. Allen works in Dr. Nilufer Ertekin-Taner's lab; their group has been looking at brain gene expression in an attempt to understand the disease better.

4. Alzheimer's isn't just dependent on inherited genetics or your environment

In the study, Allen and her team are examining variations in single nucleotides, referred to as single nucleotide polymorphisms (SNP).

These can occur when certain differences are noticed in the genetic code: On the majority of people, a specific spot may have the base cytosine, but on a small amount of people, cytosine may be replaced by the base adenine, for example.

While known variants such as this explain part of the risk for Alzheimer's disease, Allen doesn't think it explains all of it, so instead of looking at individual SNPs, she is performing an analysis that looks at the interactions between hundreds of thousands of variants throughout the genome in an effort to identify new genes that might also have an influence on Alzheimer's disease.

5. Identifying these gene variants could help us find new drug targets and open up new avenues of research

By finding new genetic variants, researchers are able to investigate them further to see if they also influence risk of Alzheimer's. This research could also advance understanding of the underlying biology of the disease, as well as finding new drug targets that can be used to treat or cure Alzheimer's.

National Science Foundation

Blue Waters is supported by the National Science Foundation through awards ACI-0725070 and ACI-1238993.