324
Audio & Video Production315
Software Development230
Automation & Workflow208
Writing & Content Creation190
Marketing & Growth179
AI Infrastructure & MLOps150
Design & Creative156
Photography & Imaging146
Data & Analytics115
Voice & Speech123
Education & Learning120
Customer Support112
Sales & Outreach115
Research & Analysis86
A Columbia and Harvard team used AI tools to redesign ribosome proteins in E. coli so part of the ribosome works without the amino acid isoleucine.
In short: Researchers used AI tools to help redesign part of a bacterial ribosome so it can work without one of life’s usual 20 amino acids.
A research team from Columbia and Harvard tested a basic question about biology, do cells really need all 20 amino acids (the small building blocks proteins are made from). Most life uses the same 20, and scientists think early life may have used fewer.
The team focused on isoleucine, an amino acid that is very similar to leucine and valine. They started with the bacterium E. coli and tried swapping isoleucine for valine in a small set of important genes. Some swaps worked, but many killed the cells or slowed their growth.
To make the test harder and more meaningful, the researchers targeted the ribosome. The ribosome is the cell’s protein factory (like a tiny assembly line that reads genetic instructions and builds proteins). They used several AI-based protein design tools to propose new protein sequences that avoid isoleucine, and they checked the likely shapes of those proteins with AlphaFold 2 (an AI system that predicts how a protein folds, like guessing how a crumpled paper will settle into a final shape).
The team eventually replaced 20 of 21 proteins in the ribosome’s small subunit with redesigned versions that used no isoleucine. After extra testing, they found one design that let cells survive with an isoleucine-free small subunit, growing at about 60 percent of the normal rate.
This work shows how today’s AI tools can speed up difficult “try and test” lab work, even for core parts of life like ribosomes. It could help scientists explore what simpler forms of biology might look like, but it also shows limits, since the redesigned cells grew more slowly and the AI tools could not clearly explain their choices.
Source: Arstechnica