Mega-analysis finds consistency in gene dysfunction in models of Alzheimer’s disease

Pictured: Dr. Paul Pavlidis. Image source: Michael Smith Laboratories/UBC.

By studying the changes in the brain in the early stage of Alzheimer’s disease (AD), researchers hope to paint a clearer picture of the biological processes leading to neurodegeneration.

An important part of understanding that landscape is making sense of the abundant genomic data that exists in the scientific literature. To find commonalities in the data on expression patterns early in the disease course, Dr. Paul Pavlidis and colleagues have conducted a large-scale analysis of ten different studies of hippocampal gene expression in mouse models of AD.

Published today in the journal eNeuro, the study found a consistent pattern of altered gene expression across several previously published studies. The researchers focused on the early stage of the disease in order to understand the set of processes underlying changes in the brain in later stages; in doing so, Dr. Pavlidis and his team were able to see trends in the data that may not have been obvious at the level of the individual study

“With mouse models one has the advantage of being able to look at the brain in early stages of disease progression. It’s important because at later stages, it’s hard to disitinguish changes in gene activity from pathology in the tissue,” said Dr. Pavlidis. “At the same time, there are multiple kinds of mouse models and there are always questions about their relevance to humans. It seemed important to combine the available data and identify any consistent effects.”

One of the most consistent changes the researchers observed were effects on genes involved in cholesterol metabolism. Cholesterol metabolism has long been considered of relevance to AD, though the mechanisms and meaning of increased expression of these genes in mice remains unclear.

Besides giving insight into AD, the study is an example of what happens when researchers make their data open.

“Without the accessibility of these data sets, we could not have established these connections,” said Dr. Pavlidis. “The consistencies in the early phase of AD models gives us new insight into what is happening at the molecular level, and can inform future research questions in a way that the individual studies can’t always do.”