The HoloFood project applies an holo-omic approach to investigate the interactions between host organisms and their gut microbiome, in relation to their diet. In particular, this project aims to improve the production of important farmed animals, such as salmon, by promoting healthier and more sustainable feed. To address this goal, several layers of omic data from salmon and associated gut microbiome were generated, analyzed, and integrated.

Host genome data generation

Two experimental trials were conducted to test the effects of adding novel sustainable feed ingredients to the diet of farmed salmon. In trial A, seaweed (Saccharina latissima), a natural symbiotic feed additive, was incorporated in the salmon diet, while in trial B, blue mussel (Mytilus edulis), a sustainable protein feed source, was added to the feed.
Host genomic data was individually generated from the DNA of 359 salmons through whole‑genome shotgun sequencing. The raw data was quality controlled, trimmed and pre‑process to remove adapters, duplicates and contaminant reads. The clean reads were then aligned to the Atlantic salmon (Salmo salar) reference genome. On average, 95.5% of the reads mapped to the reference genome, with a mean coverage of 2.6X. Mapped reads were subsequently used in the downstream genomics analysis pipeline.

Results from host genome data analysis

PCA of the genotype likelihoods generated by ANGSD, of all the samples, revealed population structuring among the individuals, with 3 main clusters potentially reflecting different paternal lineages within the cohort. The genome-wide association study (GWAS) analyses revealed no association between the host genotype and the host phenotypic traits tested in this study, such as weight, length, and specific levels of heavy metals and fatty acids. These results demonstrate that host genomic variants do not explain any of the observed variation in phenotypes measured in this study.