Profiling electric signals of electrogenic probiotic bacteria using self-attention analysis

Area of Science:

• Microbiology
• Bioelectronics
• Systems Biology

Background:

• Electrogenic probiotic bacteria produce measurable electrical signals.
• Distinguishing between bacterial species based on electrical output is challenging.

Purpose of the Study:

• To develop a method for detecting and differentiating electrical signals from Leuconostoc mesenteroides and Lactococcus lactis.
• To explore the potential of these signals as biomarkers for therapeutic efficacy.

Main Methods:

• Fabrication of a self-assembled electric circuit for signal detection.
• In vitro and in vivo (mouse intestine) experiments with probiotic inoculation and glucose.
• Analysis of electrical signal spectra using a deep learning-based self-attention mechanism.

Main Results:

• Electrical signals from L. mesenteroides and L. lactis were successfully detected on chorioallantoic membranes and in mouse intestines.
• Lactococcus lactis produced higher electricity than Leuconostoc mesenteroides in the presence of glucose.
• Distinct electrical spectra were identified for each bacterium, with unique signals highlighted by the self-attention mechanism.

Conclusions:

• The study demonstrates the feasibility of using electrical signals to differentiate between key probiotic species.
• Bacterial electrical signatures offer a novel dynamic marker for evaluating therapeutic strategies like probiotics, antibiotics, and fecal microbiota transplantation.