- AI creates a strain of E. coli with 19 amino acids, removing isoleucine from the ribosome
- This is the first organism to have less than 20 of the universal amino acids.
- The discovery supports theories about early life and opens a future for synthetic organisms in medicine.
Researchers at Columbia University, the Massachusetts Institute of Technology and Harvard University have successfully used AI to create a new version of the E. coli bacteria that eliminates one of the 20 amino acids considered the building blocks of living organisms.
A study published in Science details the importance of this discovery, which uses AI and protein engineering to create a modified strain of E. coli with only 19 amino acids, specifically dispensing with isoleucine.
This is an important milestone not only for biology, but also for AI and origins of life research. Although some organisms have already been seen to use more than 20 amino acids, scientists have never found anything with fewer than 20.
AI just allowed scientists to make an important genetic discovery
Until now, they have only been able to theorize that early primitive species may have used fewer amino acids in their genetic makeup. The discovery that follows turns that theory into a genuine possibility.
As for isoleucine being the focus amino acid within the selection of 20, its chemical similarity to leucine and valine is said to have been a major factor in the decision because it is the most substitutable.
But rather than trying to change the entire composition of the proteome (seen as the entire set of proteins within an organization), the scientists sought to harness the ribosome, which is responsible for building those proteins in the first place. As part of the experiment, the scientists were able to change the 382 isoleucine “building blocks” found in the ribosome and it continued to function as expected.
This makes it the first study of its kind, because until now scientists have only been able to edit the genetic code of bacteria, yeast and other organisms by adding amino acids, not removing them.
The experiment worked by using AI protein language models to predict alternative protein structures and offer amino acid substitutes, with the goal of preserving ribosome functions without using isoleucine.
Many of the AI-generated designs offered sequences that humans might not have designed, and AI models were able to analyze potentially successful combinations at a much higher rate than humans.
18 of the 50 E. coli strains created from the process of replacing isoleucine with alternative amino acids continued to grow normally. The next stage was to combine 21 of the rewritten ribosomal proteins into an E. coli strain, which after some further adjustments, grew (albeit more slowly than the normal, unmodified strains).
Harris Wang, a synthetic and systems biologist at Columbia University, described the complete removal of an amino acid as “almost the most difficult thing imaginable, because it is the largest and most complicated protein complex.”
What the discovery means for scientific research is that at least some core biological systems can tolerate substantial alteration of the genetic code. It also supports the argument that early life may have used fewer amino acids than modern organisms, offering new perspectives for evolutionary theories.
However, beyond evolutionary biology, the research also charts a future in which personalized synthetic organisms can play specific roles in specific tasks such as medicine and healthcare. Another conclusion is that modified organisms can become dependent on unusual chemicals not found in natural environments, thus improving biological containment.
Looking much further ahead, it is possible that AI-assisted genetic modification could help researchers engineer organisms for extreme environments, including space habitats where access to the full range of amino acids may be more limited.
Follow TechRadar on Google News and add us as a preferred source to receive news, reviews and opinions from our experts in your feeds.
