DNA extraction and virome processing methods strongly influence recovered human gut viral community characteristics

Area of Science:

• Virology
• Microbiome Research
• Bioinformatics

Background:

• Accurate characterization of the human gut virome is crucial for understanding host-microbiome interactions.
• Existing methods like multiple-displacement amplification (MDA) and total metagenomes introduce biases, complicating data interpretation and comparability.
• MDA preferentially amplifies single-stranded DNA viruses, while total metagenomes are overwhelmed by non-viral sequences.

Purpose of the Study:

• To systematically compare the biases introduced by four common human gut viral community characterization methods.
• To evaluate viral size-fraction metagenomes (viromes) prepared with and without MDA.
• To provide a framework for comparing laboratory methodologies in viral ecology studies.

Main Methods:

• Applied four methods: total metagenomes, viromes with DNase treatment, viromes without DNase treatment, and MDA viromes.
• Utilized technical triplicates for each method on a single human stool sample.
• Conducted comprehensive omic data analysis to compare viral community composition derived from each method.

Main Results:

• MDA exhibited extreme bias, with Microviridae comprising ~90% of MDA viromes versus 2% in non-MDA viromes.
• Metagenomes were enriched in temperate phages and Bacillota-phages, while viromes (both MDA and non-MDA) were dominated by Bacteroidetes-phages.
• DNase treatment had minimal impact on virome richness or composition; removing ssDNA viruses partially reduced MDA bias.

Conclusions:

• Preparatory methods significantly influence human gut virome analysis outcomes, potentially leading to divergent conclusions.
• MDA is a highly biased method that should be avoided in human gut virome studies.
• Non-MDA viromes offer a more accurate representation of the viral community, especially with sufficient DNA yields now achievable.