AI provides new insights into the ‘unknown’ bacteria on and in our bodies

 

Automation is advancing many laboratory practices in terms of sample throughout and artificial intelligence is uncovering new insights. One such example has been to reveal combinations of amino acids that feed two bacterial species.

 

Here, an artificial intelligence system was able to enable robots to conduct autonomous scientific experiments (up to 10,000 per day). The system was developed at the University of Michigan.

 

 

That artificial intelligence platform is called BacterAI and in the trial it mapped the metabolism of two microbes associated with oral health (with no baseline information to start with). The complexity faced was that bacteria consume some combination of the 20 amino acids needed to support life, but each species requires specific nutrients to grow. The researchers wished to know what amino acids are needed by the beneficial microbes in the mouth so they can promote their growth.

 

Understanding how bacteria grow is the first step toward reengineering our microbiome. Figuring out the combination of amino acids that bacteria need is not straightforward.  The 20 amino acids yield more than a million possible combinations, just based on whether each amino acid is present or not.

 

BacterAI was able to discover the amino acid requirements for the growth of both Streptococcus gordonii and Streptococcus sanguinis. To achieve this, BacterAI was required to test hundreds of combinations of amino acids per day. Within nine days, it was producing accurate predictions 90% of the time.

 

BacterAI learns by converting scientific questions into simple games that it plays with laboratory robots. The agent then distils its findings into logical rules that can be interpreted by human scientists.

 

Therefore, unlike conventional approaches that feed labelled data sets into a machine-learning model, BacterAI creates its own data set through a series of experiments. By analysing the results of previous trials, it comes up with predictions of what new experiments might give it the most information. As a result, it figured out most of the rules for feeding bacteria with fewer than 4,000 experiments. This demonstrates how automated experimentation can drastically speed up these discoveries. The team ran up to 10,000 experiments in a single day.

 

It is hoped that further AI insights will reveal more about the 90% of bacteria that humans have hardly studied.

 

The reference is:

 

Adam C. Dama, Kevin S. Kim, Danielle M. Leyva, Annamarie P. Lunkes, Noah S. Schmid, Kenan Jijakli, Paul A. Jensen. BacterAI maps microbial metabolism without prior knowledge. Nature Microbiology, 2023; DOI: 10.1038/s41564-023-01376-0

 

Posted by Dr. Tim Sandle, Pharmaceutical Microbiology Resources (http://www.pharmamicroresources.com/)