Dermatological implications of alignment-based de-hosting and bioinformatics pipelines on shotgun microbiome analysis

Area of Science:

• Microbiology
• Bioinformatics
• Dermatology

Background:

• The skin microbiome plays a vital role in skin health, and its imbalance is linked to various dermatological conditions.
• Shotgun metagenomics provides high-resolution profiling but faces challenges in dermatology due to host DNA and lack of standardized bioinformatic pipelines.
• Proprietary platforms like DRAGEN are common but limited by cost and accessibility, necessitating open-source alternatives.

Purpose of the Study:

• To evaluate an open-source, Kraken-based bioinformatics pipeline as an alternative to the proprietary DRAGEN platform for skin metagenomic analysis.
• To assess the impact of different DNA removal (de-hosting) methods on microbial and functional profiling.
• To identify a robust and customizable workflow for accurate skin microbiome research.

Main Methods:

• Systematic comparison of alignment-based de-hosting tools (Bowtie2, BWA, Rsubread) and taxonomic classifiers (Kraken2, DRAGEN).
• Utilized shotgun metagenomic data from 83 healthy individuals.
• Assessed the influence of bioinformatic choices on skin microbial composition and metabolic pathway abundance using HUMAnN 3.0.

Main Results:

• Significant variations in microbial and functional profiles were observed based on the chosen de-hosting tool and classifier.
• The combination of Bowtie2 for de-hosting, Kraken2 for classification, and HUMAnN 3.0 for functional profiling identified key sex- and age-related bacterial associations missed by other methods, including DRAGEN.
• This optimized open-source workflow demonstrated superior performance and customization potential.

Conclusions:

• Bioinformatic pipeline selection critically impacts skin microbiome analysis outcomes.
• A customizable open-source workflow (Bowtie2, Kraken2, HUMAnN 3.0) enhances reproducibility, transparency, and translational value in dermatological metagenomics.
• This approach supports the development of precision diagnostics and personalized therapies in dermatology.