AI helps scientists create first living organism with just 19 amino acids

Reviewed byNidhi Govil

2 Sources

Share

Researchers from Columbia University, MIT, and Harvard used AI to engineer a strain of E. coli bacteria with only 19 amino acids, successfully removing isoleucine from the ribosome. This marks the first organism ever created with fewer than the 20 universal amino acids, challenging fundamental assumptions about the building blocks of life and opening new possibilities for synthetic biology.

AI enables creation of first 19-amino acid organism

Researchers from Columbia University, the Massachusetts Institute of Technology, and Harvard University have achieved a scientific milestone by using AI to create a modified strain of E. coli bacteria that functions with just 19 amino acids instead of the standard 20

1

2

. Published in the journal Science, this discovery represents the first known organism to possess fewer than the universal amino acids that have long been considered fundamental building blocks of all living things. The team successfully eliminated isoleucine from the organism's genetic makeup, a feat that challenges core assumptions about the minimum requirements for life.

Source: TechRadar

Source: TechRadar

Targeting the ribosome with AI protein language models

Rather than attempting to modify the entire proteome—the complete collection of proteins within an organism—scientists focused their efforts on the ribosome, the cellular machinery responsible for building proteins

1

. The experiment used AI protein language models to predict alternative protein structures and identify amino acid substitutes that could preserve the ribosome's functions without isoleucine. These models analyzed potentially successful combinations at a pace far exceeding human capability, generating sequences that researchers might not have designed independently. The choice to remove isoleucine specifically was based on its chemical similarity to leucine and valine, making it the most replaceable of the 20 amino acids

2

.

Successful substitution across 382 building blocks

The researchers successfully replaced all 382 isoleucine building blocks found in the ribosome while maintaining its expected functionality

1

. Out of 50 E. coli strains created through this process, 18 grew normally despite the absence of isoleucine

2

. The team then combined 21 of the rewritten ribosomal proteins into a single strain of E. coli, which, after additional modifications, successfully grew—though more slowly than regular, unmodified strains. Harris Wang, a systems and synthetic biologist at Columbia University, described completely eliminating an amino acid as "almost the hardest thing you could think about, because it's the biggest, most complicated protein complex"

1

.

Implications for evolutionary biology and synthetic organisms

This breakthrough marks a departure from previous genetic modification efforts, which focused exclusively on adding amino acids to organisms rather than removing them

1

. The discovery demonstrates that core biological systems can tolerate substantial genetic code disruption, supporting theories that early, primitive species may have employed fewer amino acids in their genetic makeup

2

. Beyond evolutionary biology, the research establishes a foundation for designing customized, synthetic organisms tailored for specific tasks in medicine and healthcare. Modified organisms could become dependent on unusual chemistries not found in natural environments, improving biological containment and safety. Looking further ahead, AI-assisted genetic modification could help design organisms for extreme environments, including space habitats where access to the full range of amino acids may be limited

1

.

Today's Top Stories

TheOutpost.ai

Don’t drown in AI news. We cut through the noise - filtering, ranking and summarizing the most important AI news, breakthroughs and research daily. Spend less time searching for the latest in AI and get straight to action.

Instagram logo
LinkedIn logo
Youtube logo
© 2026 TheOutpost.AI All rights reserved