// NATURE NEWS — SPAZIO & SCIENZA
AI agents run this automated biology lab that sequences DNA and more
Heidi Ledford reports for Nature from Seattle, Washington.
Search author on:
PubMed
Google Scholar
A self-driving lab named AutoDNA can work out the constituent ‘letters’ in a snippet of DNA.Credit: Turtle Rock Scientific/Science Source/Science Photo Library
To say that someone has ‘good hands’ is high praise in a molecular biology laboratory, where researchers must deftly handle delicate glassware, pipette samples and more. But in an unusual genetics lab at Shanghai Jiao Tong University in China, a team of robots driven by artificial intelligence is designing and doing experiments with no hands to speak of.
The facility, described today1 in Nature, is one of the latest steps towards ‘self-driving’ labs that are guided by AI, with minimal input from human scientists. With samples moved around by robotic arms, the Shanghai lab can operate more than 20 types of instrument, complete thousands of operations in a single project and do experiments that involve synthesizing and sequencing DNA, among other tasks.
An experiment is designed and executed from start to finish by a team of AI agents, following a simple prompt from a human operator. After that, human hands are needed only to refresh reagents and to service hardware.
The system, called AutoDNA, is remarkable both for its degree of autonomy and because it can readily incorporate a diverse range of lab equipment, says Milad Abolhasani, a chemical engineer at North Carolina State University in Raleigh. “The future of science is going to involve a lot of autonomous labs.”
Labs have long incorporated automation into their workflows, using machines to conduct specific aspects of experimental protocols, such as mixing and heating chemicals. Now, some researchers are hoping to harness increasingly sophisticated AI tools to create self-driving labs, in which an AI system designs and executes complex experiments, troubleshoots and tweaks protocols, and orchestrates activities across an entire lab that is chock-full of disparate equipment.
Designing such labs for molecular biology is challenging because of the field’s complex workflows, which require assorted equipment and high precision, says Haibo Chen, a computer scientist at Shanghai Jiao Tong University and a co-author of the report. “In many existing systems, human experts are still required to configure the instruments,” he says. “AI is mostly added to handle just one part of the process.”
To create a more fully autonomous lab, Chen and his colleagues decided to construct it from the start with AI in mind, rather than adding AI capabilities to a pre-existing system.