Posted in | News | Medical Robotics

Machine Learning Used to Interpret Genetic Influence over Behavior

Mice scamper around while searching for food, but genetics may be the hidden hand regulating these meandering movements. Scientists at the University of Utah Health are using machine learning to derive connections between genetic controls that outline the incremental steps of instinctive and learned behaviors. The results of the research can be seen in the August 13th online issue of Cell Reports.

Genetics control behavioral sequences that are building blocks for complex behavior patterns. (Image credit: University of Utah Health)

Patterns of complex behavior, like searching for food, are composed of sequences that feel random, spontaneous and free. Using machine learning, we are finding discrete sequences that are reproduced more frequently than you would expect by chance and these sequences are rooted in biology.

Christopher Gregg, Ph.D., Study Senior Author and Assistant Professor in Neurobiology and Anatomy, University of Utah Health

The team is diving into the new territory of behavioral sequencing.

“We are trying to understand the architecture of complex behavior and how genetics shape these patterns,” said Gregg.

The study supports the concept that complex behavior is made up of an assortment of finite “building blocks” the authors refer to as behavioral modules and that genetics are regulating the progression of these building blocks to develop various behavioral patterns.

The team tested 190 mice with differences in their age and genetics as they moved from their home into a uniquely built “arena” to assess the set of behavioral sequences expressed while searching for food. In the hunt for food, mice display behaviors that necessitate a number of neural systems to control seeking-behaviors, preservation, navigation, anxiety, hunger, reward, attention, satiety, and memory. The new techniques exposed that different genetic and age effects impact different sequences.

Most species have a home range and their behaviors are structured around this home range. We were able to identify reproduceable behavioral sequences and use this information to understand the complex patterns over time.

Christopher Gregg, Ph.D., Study Senior Author and Assistant Professor in Neurobiology and Anatomy, University of Utah Health

The team divided round trips from home to a food source and back into a sequence of over 5,600 mouse actions. Layered within these actions are extra information, such as distance traveled, velocity, gait pattern, and locations visited. With the help of machine learning, they assessed this information and recognized 71 reproducible behavioral sequences that are the fundamental building blocks for more complicated behavior patterns.

The changeover from one “building block” to the next suggests a mechanistic relationship that produces specific foraging behaviors that lessen energy expenditure, predation risk, and caloric intake. Furthermore, the algorithm was able to identify impulsive responses that are unique to some mice.

Gregg believes this method is adequately sensitive to pick up a mutation in the copy of one gene. To establish this point, his team concentrated on hunting behaviors in mice with a mutation in an imprinted gene, Magel2, which is connected to autism. For instance, when the mother’s copy is turned off, the father’s copy is turned on. In this situation, it was broadly believed that the mother’s copy was silent and did not impact the offspring. But this is not so.

“What was exciting to us was we were able to detect significant effects on behavior from a single mutation in only the mother’s gene copy,” Gregg said.

Presently, the research has only investigated the building blocks of foraging behavior in lab mice. Gregg thinks the methodology could be applied to comprehend the foundation of other complicated behavior patterns and learn the particular genomic elements that mold behaviors leading to disease in humans, including addiction, obesity, anxiety, fear, and psychiatric disorders.

By deconstructing really complex, seemingly spontaneous behaviors, we were able to detect things that weren’t observable in other studies. If there is a mutation that causes disease in people we hope to use this method to map it to specific modules [aka behavior building blocks] to learn how genes contribute to shaping particular behavior patterns.

Christopher Gregg, Ph.D., Study Senior Author and Assistant Professor in Neurobiology and Anatomy, University of Utah Health


Tell Us What You Think

Do you have a review, update or anything you would like to add to this news story?

Leave your feedback
Your comment type