Mapping the brain's traffic

Carnegie Mellon University researchers are working to learn how neurons in the brain work together to understand and treat neurodegenerative diseases such as Alzheimer's.

Using novel engineering models, methods and software, Ge Yang, associate professor of biomedical engineering and computational biology, and Jessica Zhang, professor of mechanical engineering, are studying how essential materials, such as chemical signals and cell parts made in the nucleus, are transported within the complex geometry of neurons and creating simulations that represent this transport system. For their efforts, they were awarded a National Science Foundation (NSF) grant to continue their research for three years.

Neurons come in many shapes and sizes, but each has the same foundational structure. The axon is a thin, long wire-like structure that transports information, signals and materials from the cell body to communicate with other neurons. Similar to how an urban center distributes materials throughout metropolitan areas and collects garbage and recycling from these areas, a neuron sends out and collects materials and signals. This transport system is critical for brain function and ultimately human survival and is at the center of Yang and Zhang's research.

The simulation result in a mouse cerebellum Purkinje neuron showing the dynamic material transport process for 340 seconds.

"This process is literally a matter of life and death for a neuron," Yang said. "If this process shuts down, then the neurons will die. For example, with Alzheimer's disease, the brain is smaller size-wise, but it also has many holes. These holes are made due to massive death of neurons. There are all kinds of theories about Alzheimer's, and one of the theories which has received fairly strong experimental data support is that there's a traffic problem."

To better understand how the complex geometry of different neurons affect how data is transported and distributed, Yang and Zhang are developing computer simulations of how materials are transported through the neuron geometry.

"We're using a very advanced method called isogeometric analysis," Zhang said. "The neurite tree of a neuron is very complex - we use more than 1,000 processors for the simulations."

Isogeometric analysis (IGA) is a computational method that allows numerous models to be designed and tested from a data set all at once. Using IGA, Yang and Zhang simulate the flow of materials inside a neuron. The warmer the color, the higher the density of flow. As the materials flow from branch to branch away from the cell body, the color cools. At branch junctions, where there is more traffic, there is higher flow and thus a warmer color.

Examples of complex neuron geometry

"Jessica's lab is using the supercomputer in the Pittsburgh Supercomputing Center because the calculations are very complex," Yang said. "When we do this, it involves millions of units in the calculations. We're really pushing the limits of the technology right now. If the computer simulation is done properly, it is going to have a very large impact on people's understanding of the neuron's geometry."

In 2016, Yang and Zhang received seed support from the Department of Mechanical Engineering, which allowed them to take on a Ph.D. student, Angran Li, who has worked on developing software and simulations for this project.

Overall, Yang and Zhang have two main goals. First, they want to better understand the transport process of neurons, which can provide new insight on the development of neurodegenerative diseases. Second, when it comes to neurodegenerative disease, they hope to learn more about how drug delivery can be effectively distributed in the complex structure of neurons.

"It's a fundamental scientific and engineering question," Yang said. "The long-term goal is to develop a whole theory on how material is transported. This complex structure, with the physiology of neurons — it has a fundamental impact on life."