Machine learning-based prediction of BV-relevant probiotic functional potential in vaginal-derived Lactobacillus...

Area of Science:

• Microbiology
• Biotechnology
• Genomics

Background:

• The vaginal microbiota is crucial for female reproductive health.
• Current Lactobacillus probiotics are mainly for intestinal health, with limited focus on vaginal applications.
• Traditional screening methods for probiotics are costly and labor-intensive.

Purpose of the Study:

• To develop an efficient method for identifying beneficial Lactobacillus crispatus strains for female reproductive health.
• To leverage machine learning for rapid analysis of omics data to find candidate probiotics.
• To establish a predictive pipeline for assessing the functional potential of vaginal Lactobacillus crispatus.

Main Methods:

• Isolated 639 Lactobacillus crispatus strains from healthy women's vaginal secretions.
• Conducted in vitro assays (growth, acidification, lactic acid, H2O2, antimicrobial activity) on 67 representative strains.
• Developed a weighted scoring system and a machine learning pipeline (VLCPredictor) integrating genomic features.

Main Results:

• Identified three Lactobacillus crispatus strains with superior in vitro performance.
• VLCPredictor pipeline demonstrated capability in scoring functional potential of vaginal Lactobacillus crispatus.
• The study established an efficient framework for probiotic development.

Conclusions:

• The developed framework offers an efficient approach for identifying and applying Lactobacillus crispatus for female reproductive health.
• This method can aid in maintaining vaginal microbial equilibrium and preventing gynecological infections.
• The findings support the development of targeted probiotics for women's health.