(Yicai) Oct. 15 -- Chinese researchers have developed an artificial intelligence-driven platform to accelerate the selection of high-performance microbial clones, a breakthrough that can be used in fields such as medicine, bioengineering, and industrial production.

A research team from the Qingdao Institute of Bioenergy and Bioprocess Technology under the Chinese Academy of Sciences created the Digital Colony Picker (DCP) to transform a once slow, labor-intensive process into a fast and automated workflow, according to a recent study published in Nature Communications.

The DCP is a system that uses a microfluidic chip with 16,000 addressable microchambers to isolate single cells and monitor their growth into micro-colonies.

Specifically, the platform’s built-in AI engine performs real-time time-lapse analysis of brightfield and biosensor signals to measure growth kinetics and metabolite production. Once target colonies are identified, they are exported via a laser-induced bubble technique as droplets directly into standard culture plates. This contact-free transfer reduces cross-contamination and preserves cell viability.

To demonstrate its efficiency, the researchers used DCP to select Zymomonas mobilis, a Gram-negative bacterium known for ethanol fermentation in biomass energy production. They identified a strain that grew 77 percent faster under 30 g/L potassium lactate and achieved nearly 20 percent higher lactate titer than the control group after one round of screening.

The study noted that the DCP’s design solves key bottlenecks in droplet- and plate-based screening. Gas-phase separation prevents cross-contamination, rapid medium exchange supports long-term assays, chip-wide indexing enables panoramic analysis of about 800 colonies per minute, and push-button, contact-free export ensures efficient collection and robust growth.

The researchers added that the platform could also be applied to adaptive evolution studies, functional gene discovery, and phenotype screening across diverse microbial species.

Editor: Emmi Laine