Bacteria Might Be Used As Live Robot Brains

Scientists published a report this week in which a computer model was used to investigate the possibility of employing bacterial colonies to control mechanical devices like robots. The results were remarkably positive. Read on to find out how this new “living brain” development will impact everything from ecology to medicine.

Honda's ASIMO robot: one day robots like this may be biological-mechanical hybrids with bacterial "brains". Image taken by Gnsin at Expo 2005, in Japan.Honda's ASIMO robot: one day robots like this may be biological-mechanical hybrids with bacterial "brains". Image taken by Gnsin at Expo 2005, in Japan.

The research was conducted by Virginia Tech assistant professor Warren Ruder in collaboration with biomedical engineering graduate student Keith Heyde and published this month in the journal Scientific Reports. Their idea came from a previous study that showed the mating habits of fruit flies could be manipulated through bacterial intervention. This initial study of an inanimate object controlled by a microbiome was carried out in silico – scientists’ favorite term for theoretical research conducted on a computer. In their model system, three components were simulated all of which can be described by widely accepted equations and criteria lending credence to the results obtained:
1.    Engineering gene circuits in bacteria (E. coli were chosen due to their ubiquity and the degree to which we understand their genome).
2.    Microfluid bioreactors.
3.    Robot movement.

Escherichia coli: electron microscope image taken at Rocky Mountain Laboratories, NIAID, NIH.Escherichia coli: electron microscope image taken at Rocky Mountain Laboratories, NIAID, NIH.

The computer model then had only to communicate how these three components could be usefully designed to interact together. The bacteria were engineering to change between green and red depending on their diet. To take advantage of this signaling system, the robot host was equipped with sensors and a microscope to detect bacterial color changes. In response, the robot changed its velocity in response to both the color and pigment intensity of the bacteria. Though it seems (and is) very simplistic, some remarkable behavior emerged in the model, independent of the researchers’ programming. For example, when approaching a food source, the robot paused when it got close and then pounced – a trait typically seen in higher order mammals as they stalk their prey.

Aside from just serving as a central control “brain” in robotics, there are a number of possible advances that will stem from the integration of bacteria and machine. In particular, the researchers hope to use this model system to develop of greater understanding of bacterial interactions on a broad number of fronts; for example, the connection between livestock and soil bacteria. Perhaps of greater interest is the way this research could help us recognize how bacteria in the human body influences our physical and mental health. Indeed, an increasing body of research suggests that bacteria can actually affect our behavior and mood, and understanding how and why this occurs could lead to treatments for currently unmanageable mental illnesses.

Check out Dr. Ruder's explanation in this video:

With the model offering firm confirmation that microbiome-machine hybrids are feasible, the next step for Ruder is the fabrication of a working robot. He plans to equip the machine with small scale fluorescent microscopes coupled with bacteria specially engineered in his lab to exhibit variable traits detectable by such a system. He also envisions other applications for this system like robotic drones that can be deployed to oil spills where they’ll release bacterial oil scrubbers. No matter what the eventual application, the idea of circumventing the need for developing a ground-up synthetic brain by utilizing an already living entity is brilliant and this research is likely just a first small step toward a major new field in robotics.

Via Virginia Tech and Scientific Reports.