Global Microbiome Congress

Abstract

Reproducible metagenomic insights depend on harmonised laboratory workflows, yet cross-platform benchmarks that span the entire wet-lab pipeline are still limited. We created an open, 106 GB multi-omic resource that quantifies how DNA extraction, primer design and sequencing technology interact to shape gut-microbiome profiles. Using the dog gut—an established proxy for the human microbiome—we:

1. Deep-benchmarked a reference stool sample with five extraction kits, eight library chemistries and four sequencing platforms (short- and long-read WGS, full-length 16S).
2. Profiled 40 longitudinal canine samples processed in parallel with two widely used extraction kits to capture population-scale effects.
3. Compared three full-length 16S primer sets (standard ONT, degenerate ONT, PacBio) across synthetic mocks, human stool and canine stool to expose primer-driven taxonomic distortions.

Cross-species analysis of the primer-comparison revealed that in some cases (e.g., Bifidobacterium, Ruminococcus), the human fecal reference mock community’s relative abundance aligned more closely with the canine than the human sample, reaffirming reports that the canine gut microbiome more closely mirrors the human counterpart than those of traditional rodent models. Extraction chemistry explains the largest share of compositional variance; degenerate primers rescue under-represented clades such as Bifidobacterium; and long-read WGS helps disentangle platform-specific biases that confound short-read studies.

All raw reads, metadata and R scripts are publicly released, offering a plug-and-play benchmark for validating human metagenomic pipelines, optimising cross-study comparisons and designing translational dog–human projects. By delivering a rigorously controlled, species-bridging testbed, we provide actionable guidance for standardising gut-microbiome workflows that will accelerate reproducible discoveries in human health research.

Biography