First, artificial intelligence trumped expert chess players at their own game. Then came Watson, a computer system that famously beat Jeopardy! champions Brad Rutter and Ken Jennings. Now, researchers are putting artificial intelligence to work to automate biological research—-specifically the reverse engineering of metabolic networks from experimental data.
A team of scientists from Vanderbilt University, Cornell University, and CFD Research Corp. have shown that a computer can take raw experimental numbers from a biological data and derive equations from it that describe how the system functions. The modelling used in the research is said to be one of the most complex scientific modeling problems that artificial intelligence has solved completely from scratch.
Check out the announcement from Vanderbilt University:
The “brains” of the system, which [Vanderbilt professor John P. Wikswo] has christened the Automated Biology Explorer (ABE), is a unique piece of software called Eureqa developed at Cornell and released in 2009. [Michael Schmidt and Hod Lipson at the Creative Machines Lab at Cornell University] originally created Eureqa to design robots without going through the normal trial and error stage that is both slow and expensive. After it succeeded, they realized it could also be applied to solving science problems.
One of Eureqa’s initial achievements was identifying the basic laws of motion by analyzing the motion of a double pendulum. What took Sir Isaac Newton years to discover, Eureqa did in a few hours when running on a personal computer.
In 2006, Wikswo heard Lipson lecture about his research. “I had a ‘eureka moment’ of my own when I realized the system Hod had developed could be used to solve biological problems and even control them,” Wikswo said. So he started talking to Lipson immediately after the lecture and they began a collaboration to adapt Eureqa to analyze biological problems.
“Biology is the area where the gap between theory and data is growing the most rapidly,” said Lipson. “So it is the area in greatest need of automation.”
Wikswo believes that artificial intelligence could potentially be harnessed to generate and analyze biological data to such a degree that it could predict the behavior of biological systems under a variety of conditions.
[Wikswo also] maintains that this approach will give scientists the ability to control biological systems even if [the researchers] can’t completely explain how they work, and this capability can provide the basis for the development of significantly improved drugs and other therapies.
According to Cornell professor Hod Lipson, the researchers might need to create another program to translate the meaning of the equations that the Eureqa program comes up with.
This this video from a couple of years ago explains how the Eureqa software derived the fundamental equations of motion from observations of a double pendulum.
Top image: The microformulator pictured enables the biological experiments to be performed without human intervention. Image credit: Wikswo Lab.
Abstract in Physical Biology: Automated refinement and inference of analytical models for metabolic networks
Press release: Robot biologist solves complex problem from scratch