AI-RACS: Innovative tool for mining aluminum-tolerant strains in acidic soils

Researchers from the Single-Cell Center at the Qingdao Institute of Bioenergy and Bioprocess Technology of the Chinese Academy of Sciences (CAS), together with collaborators, have developed an artificial intelligence-assisted Raman-activated cell sorting (AI-RACS) system. This innovative system has automated the isolation and functional analysis of aluminum-tolerant microorganisms (ATMs) from acidic soil, marking a shift from manual, labor-intensive procedures to high-throughput automated workflows.

This study was published in Analytical Chemistry.

Microbiomes—dynamic communities of microorganisms—offer untapped potential for advancing biotechnology and environmental sustainability. However, their complexity poses challenges for the isolation and detailed study of specific functional microbes.

To address this issue, the AI-RACS system integrates optical tweezers, single-cell Raman spectroscopy (SCRS), and artificial intelligence. This integration enables the precise identification, sorting, and collection of single cells, transforming microbial single-cell research from low-throughput manual operations to high-throughput automated workflows.

The researchers utilized the RACS-Seq/Culture instrument to identify and sort ATMs from acidic soil samples. By employing SCRS to assess cellular metabolic activity under aluminum stress, the researchers successfully identified and isolated 13 aluminum-tolerant strains, including Burkholderia spp., Rhodanobacter spp., and Staphylococcus aureus. These strains exhibited higher metabolic activity compared to those identified through traditional cultivation methods. The use of SCRS as a quantitative biomarker allowed the researchers to pinpoint and categorize metabolically active microbes with unmatched precision.

"AI-RACS allows us to uncover how ATMs thrive in toxic red soils, providing new perspectives on microbial survival and soil health restoration," said Prof. LIANG Yuting, corresponding author of the study, from the Institute of Soil Science of the CAS.

"Our goal is to develop a system that automates single-cell analysis while improving precision and throughput needed for studying complex microbial communities," said Dr. DIAO Zhidian, first author of the study, from the Single-Cell Center. "This system enables researchers to explore microbiomes under near in situ conditions with high efficiency."

The AI-RACS system opens up new possibilities in fields such as resource recovery, environmental management, and industrial biotechnology.