Microbiome

Microbial genomes as genetic mediators: a comparative 16S rRNA and shotgun metagenomic study in Iberian pigs

---

Abstract

Advances in metagenomics allow integration of microbial profiles into animal breeding models, offering a novel perspective on how host-microbiome interactions influence phenotypic variability. This study compared 16S rRNA gene sequencing and shotgun metagenome sequencing (WGS) in their ability to capture microbial and genomic components of meat quality and carcass traits in Iberian pigs. 162 Iberian pigs were raised under standardized management at the Iberian Testing Center (Spain). Fecal samples collected before slaughter (≈365 days) were analyzed using nf-core/ampliseq (16S) and SqueezeMeta (WGS). After quality control, 275 genera, 212 species and 4527 KEGG Orthologs-KOs (WGS) and 154 genera, 44 species and 1775 KOs (16S) were retained. Animals were genotyped using the Illumina GGP PorcineHD BeadChip (35,285 SNPs post-QC). Phenotypes included fatty acid (FA) composition of longissimus thoracis muscle and carcass and primal-cut yields (ham, loin, shoulder). A three-step Bayesian framework (BGLR, R) was applied to disentangle microbial, genetic, and microbe-mediated components of trait variation. First, microbiability was estimated using a mixed model with random additive genetic and microbial effects. Second, microbe-specific effects were obtained using BayesC regression including additive genetic effects and the microbial abundance matrix to estimate regression coefficients. Third, each microbe was modeled as a quantitative trait to estimate its additive genetic variance, allowing estimation of the fraction of trait genetic variance mediated through each microbe, and ranking taxa according to their contribution. Both sequencing approaches revealed a substantial microbial contribution to phenotypic variation.At genus level, microbiabilities were similar between methods for both FA (16S: 0.20-0.23; WGS: 0.22-0.24) and carcass traits (16S: 0.14-0.32; WGS: 0.14-0.20). In contrast, at species level, WGS yielded relevantly higher estimates than 16S for FA traits (0.19-0.27 vs. 0.09-0.11) and carcass traits (0.14-0.20 vs. 0.07-0.11), except for ham yield, where estimates were identical (16S: 0.21; WGS: 0.19). Conversely, at the functional level (KOs), microbiability estimates were comparable (16S: 0.16-0.33; WGS: 0.15-0.32), with 16S slightly higher across most traits. Importantly, microbe-specific regression coefficients showed marked heterogeneity and uneven contributions, with few taxa explaining large proportions of the microbiome-mediated genetic variance. At genus level, Prevotella, Lactobacillus, and Streptococcus accounted for 37-93% of the total microbiome-mediated genetic variance depending on the trait. At species level, Escherichia coli, Lactobacillus amylovorus, and Streptococcus gallolyticus explained 22-93% depending on the phenotype. Overall, the similarity of microbiability estimates between 16S and WGS at the genus and functional levels indicates that both approaches capture microbial contributions to similar extents at these resolutions. At the species level, however, WGS yields substantially higher estimates, supporting stronger evidence for species-associated variance components, whereas consistently lower 16S-based estimates suggest limited ability to capture species-level microbial effects. Accordingly, WGS is required to detect species-level microbial effects, whereas 16S remains adequate for analyses targeting genus- or functional-level signals.

Keywords: 2026

How to Cite:

Azouggagh, L., Casto Rebollo, C., Varona, L., Casellas, J., Negro, S. & Ibañez-Escriche, N., (2026) “Microbial genomes as genetic mediators: a comparative 16S rRNA and shotgun metagenomic study in Iberian pigs”, World Congress on Genetics Applied to Livestock Production Digital Archive 2026(1): 2285634. doi: https://doi.org/10.31274/wcgalp.23762

Rights: 1