© Chokniti Khongchum/Pexels
In the field of molecular biology, AI has made giant strides and is completely redefining experimental protocols. Today, we are witnessing the emergence of autonomous laboratories (like the A-Lab), driven entirely by robots and AI models. It's a whole new way to approach scientific research and accelerate it.
The emergence of autonomous laboratories
A functional autonomous laboratory combines two cutting-edge technological packages :
- Robotic equipment.
- Models machine learning with the ability to design experiments and analyze the results.
An approach completely revolutionary which accelerates the scientific process and generates innovative solutions. Héctor García Martín is a physicist and synthetic biologist at Lawrence Berkeley National Laboratory. He explains: “c‘is cutting-edge work“. These labs “completely automate the entire protein engineering process”.
It’is a special field of biotechnology focusing primarily on the creation or modification of proteins. Its applications are rather diverse: medical, industrial or pure research. It necessarily involves the manipulation of protein structure; a process which can be rather tedious if we confine ourselves to traditional methods.
Indeed, to determine which protein presents the best performance, it is necessary to carry out a very large number of manipulations and tests. A job that can quickly become monotonous and repetitive. Automating all of these processes using AI and robotics is a huge time saver for researchers.
The great power of a simplified AI model
At the University of Wisconsin-Madison, the team research of Philip Romero (molecular geneticist specializing in proteins) has developed a system based on a simple AI model. This is capable of linking a protein to its function and quickly proposing the appropriate sequence modifications to improve it.
According to Romero himself, he explains that this model allows “set and forget“ ;. To understand: “we give the instructions to the model, and it's finished, no need to worry about it anymore“. The model is autonomous and drives almost aloneenzyme reconfiguration experiments. By this term we designate the entire process of modifying or redesigning enzymes intended to improve their performance or to create new specific functions.
Their model sends the sequences of proteins he composed to laboratory equipment which then makes the proteins, tests them, and the process is repeated. Thanks to this approach, the team succeeded in making certain metabolic enzymes (glycoside hydrolases) more resistant to heat. After 20 experimental cycles only, they achieved very convincing results.
Future challenges and collaboration with humans
Even if this wave of laboratory automation opens up a vast and very promising horizon of possibilities, adjustments are still necessary to adapt this mode of operation to other fields< /strong> than the genius of proteins.
Huimin Zhao, a biologist at the University of Illinois Urbana-Champaign, says making proteins more heat stable is a relatively simple process. On the other hand, adapting a stand-alone laboratory to modify enzymes in another way is not yet an obvious task.
The goal of these autonomous laboratories is absolutely not to replace the hand of man. Rather, they are designed to reduce repetitive and laborious tasks as much as possible, so that researchers can concentrate instead on the creative side of their profession. Jacob Rapp, co-author of the paper published in Nature on which this article is based, expresses this very well to his colleagues: & ;#8220;we don't replace humans, we replace the boring parts so you can focus on the interesting aspects of your engineering work “.
Automation and AI herald a truly exciting future for molecular biology, and for scientific research at large. Even if autonomous laboratories are not yet applicable to all fields, see that they are already capable of carrying out a titanic work in so little time is already great. A collaboration between man and machine which already promises to be a game changer in the field of research.
- Autonomous laboratories, managed by robots and AI systems are increasingly emerging.
- In the field of protein engineering, automation is already making wonders.
- For the moment, these automation processes are not necessarily possible in all areas.
📍 To not miss any news from Presse-citron, follow us on Google News and WhatsApp.