MG-MLST: Characterizing the Microbiome at the Strain Level in Metagenomic Data

Area of Science:

• Microbiology
• Genomics
• Bioinformatics

Background:

• The human microbiome is crucial for physiology, but strain-level composition remains largely uncharacterized.
• Strain-level microbial differences are increasingly recognized for their role in disease associations.
• Existing methods for strain identification often require extensive sequencing and reference genomes.

Purpose of the Study:

• To develop and validate a novel method, metagenomic multi-locus sequence typing (MG-MLST), for determining microbial strain-level composition.
• To assess the accuracy of MG-MLST using simulated communities and real-world clinical samples.
• To investigate strain-level differences in the skin microbiome, particularly in relation to acne.

Main Methods:

• Developed MG-MLST by combining high-throughput sequencing with multi-locus sequence typing (MLST).
• Tested MG-MLST on simulated microbial communities and clinical skin samples.
• Validated MG-MLST results against 16S rRNA clone libraries and metagenomic shotgun sequencing.

Main Results:

• MG-MLST accurately predicted strain populations in simulated communities.
• The method yielded consistent strain composition results compared to established techniques.
• Identified specific microbial strains (RT2/6) associated with healthy skin, differentiating from acne-associated microbiomes.

Conclusions:

• MG-MLST offers a quantitative approach to analyze microbiome strain diversity and richness.
• This method enables the detection of critical strain-level differences between groups.
• MG-MLST is a valuable tool for microbiome studies, especially in understanding microorganism-related diseases.