Revealing the networks that guide brain function

June 11, 2015

We often say that no man is an island, and when it comes to high-level behaviors like vision, language, episodic memory and spatial attention, the same can be said of the brain’s regions. Rather than a patchwork of separated areas, functional MRI (fMRI) studies provide evidence of “functional networks” in the brain, with as many as 15 to 20 disparate regions active at the same time. These functional networks are also disrupted in several brain diseases and disorders, including Alzheimer’s disease.

Though fMRI gives ample evidence of functional networks, understanding the molecular mechanisms that allow those brain regions to work together has been challenging. This is in part because correlating two very different kinds of information—fMRI images and genetic data—is nuanced and difficult to control.

Researchers at Stanford University, in collaboration with the Allen Institute for Brain Science and the IMAGEN Consortium, recently used three Allen Institute resources to correlate gene expression and connectivity data with resting state fMRI. Their illuminating results show how genes that control for ion channel and synapse function, both crucial for sending neural signals, underlie the functional networks we see in fMRI images. The work is published this week in the journal Science.

The researchers began with eight-minute fMRI scans of 15 healthy adults to identify key resting state functional networks. They then turned to the Allen Human Brain Atlas to search for genes that were more highly expressed within the networked regions than anywhere else in the brain.

“We wanted to discover which genes could be underlying the functional networks we observe in fMRI data,” says Michael Hawrylycz, Ph.D., Investigator at the Allen Institute, who advised on the study. “Differences in gene expression in the human neocortex can be subtle, so we used the detailed gene expression data from the Allen Human Brain Atlas to define ‘correlated gene expression networks,’ which show how genetically similar different regions of the networks are to each other compared to the rest of the brain.”

The list of 136 genes the researchers generated contained many genes crucial for propagating signals, including ion channels which allow for signals to be transmitted down the length of a neuron, and synapses where cells pass signals from one to the next.

The team used several other databases to confirm and bolster their results, including the Allen Mouse Brain Atlas, the Allen Mouse Brain Connectivity Atlas, and the IMAGEN Consortium database which contains genetic profiles and cognitive tests of adolescents.

“Showing that we can correlate fMRI images with gene expression provides useful data that can help us better understand the brain’s functional networks,” says Hawrylycz. “Hopefully this work will eventually also shed light on what goes wrong in diseases when these functional networks are disrupted,” says Hawrylycz.