AI-Driven Robotic Experiments: Revolutionizing Scientific Discovery in Medicine, Agriculture, and Environmental Science

Artificial Intelligence Enables Robots to Conduct Thousands of Scientific Experiments per Day

An artificial intelligence system has revolutionized the field of scientific experimentation by allowing robots to autonomously perform up to 10,000 experiments per day. This breakthrough has the potential to greatly accelerate scientific discoveries in various fields such as medicine, agriculture, and environmental science.

Mapping Microbial Metabolism with BacterAI

In a recent study published in Nature Microbiology, a team of researchers led by a professor from the University of Michigan introduced an AI platform called BacterAI. They used this platform to map the metabolism of two microbes associated with oral health, even though they had no initial information to start with.

The U-M team aimed to understand the amino acid requirements of these beneficial mouth microbes in order to promote their growth. Bacterial growth is influenced by the consumption of specific nutrients, and uncovering the amino acid combination required by different species of bacteria is a complex task. With 20 amino acids yielding over a million possible combinations, finding the right formula can be challenging. However, BacterAI successfully discovered the amino acid requirements for the growth of Streptococcus gordonii and Streptococcus sanguinis.

Accelerating Discoveries through Autonomous Experimentation

Unlike traditional methods that rely on labeled data sets, BacterAI creates its own data set through a series of experiments, enabling it to make predictions for future experiments. By analyzing the results of previous trials, BacterAI determines the most informative experiments to pursue. This approach allowed the AI system to uncover the majority of bacterial feeding rules in less than 4,000 experiments.

Dr. Paul Jensen, assistant professor of biomedical engineering at U-M, compares BacterAI’s learning process to that of a child learning to walk. Just as a child stumbles and makes mistakes before mastering walking, BacterAI learns and improves through trial and error. With each passing day, the AI system becomes more knowledgeable and efficient in its predictions.

Expanding the Reach of Automated Experimentation

The potential of autonomous experimentation extends beyond microbiology. Researchers in any field can utilize the power of AI to design experiments that can be solved through trial and error. This approach has the potential to greatly accelerate the pace of scientific research in various domains, leading to exciting discoveries and advancements.

Adam Dama, lead author of the study and a former engineer in the Jensen Lab, emphasizes the positive impact of focused AI applications like BacterAI. He believes that such technology will significantly enhance everyday research efforts. With the ability to run up to 10,000 experiments in a single day, the possibilities for scientific exploration are vast.

The research for this project was supported by the National Institutes of Health and funded in part by NVIDIA.

Source link

Stay in the Loop

Get the daily email from AI Headliner that makes reading the news actually enjoyable. Join our mailing list to stay in the loop to stay informed, for free.

Latest stories

You might also like...