Summary From ontogenesis to homeostasis, the phenotypes of complex organisms are shaped by the bidirectional interactions between the host organisms and their associated microbiota. Current technology can reveal many such interactions by combining multi-omic data from both hosts and microbes. However, exploring the full extent of these interactions requires careful consideration of study design for the efficient generation and optimal integration of data derived from (meta)genomics, (meta)transcriptomics, (meta)proteomics, and (meta)metabolomics. In this perspective, we introduce the holo-omic approach that incorporates multi-omic data from both host and microbiota domains to untangle the interplay between the two. We revisit the recent literature on biomolecular host-microbe interactions and discuss the implementation and current limitations of the holo-omic approach. We anticipate that the application of this approach can contribute to opening new research avenues and discoveries in biomedicine, biotechnology, agricultural and aquacultural sciences, nature conservation, as well as basic ecological and evolutionary research.

Fast optimisation of farming practices is essential to meet environmental sustainability challenges. Hologenomics, the joint study of the genomic features of animals and the microbial communities associated with them, opens new avenues to obtain in-depth knowledge on how host-microbiota interactions affect animal performance and welfare, and in doing so, improve the quality and sustainability of animal production. Here, we introduce the animal trials conducted with broiler chickens in the H2020 project HoloFood, and our strategy to implement hologenomic analyses in light of the initial results, which despite yielding negligible effects of tested feed additives, provide relevant information to understand how host genomic features, microbiota development dynamics and host-microbiota interactions shape animal welfare and performance. We report the most relevant results, propose hypotheses to explain the observed patterns, and outline how these questions will be addressed through the generation and analysis of animal-microbiota multi-omic data during the HoloFood project.

Metagenomic datasets of host‐associated microbial communities often contain host DNA that is usually discarded because the amount of data is too low for accurate host genetic analyses. However, genotype imputation can be employed to reconstruct host genotypes if a reference panel is available. Here, the performance of a two‐step strategy is tested to impute genotypes from four types of reference panels built using different strategies to low‐depth host genome data (≈2× coverage) recovered from intestinal samples of two chicken genetic lines. First, imputation accuracy is evaluated in 12 samples for which both low‐ and high‐depth sequencing data are available, obtaining high imputation accuracies for all tested panels (>0.90). Second, the impact of reference panel choice in population genetics statistics on 100 chickens is assessed, all four panels yielding comparable results. In light of the observations, the feasibility and application of the applied imputation strategy are discussed for different species with regard to the host DNA proportion, genomic diversity, and availability of a reference panel. This method enables leveraging insofar discarded host DNA to get insights into the genetic structure of host populations, and in doing so, facilitates the implementation of hologenomic approaches that jointly analyze host and microbial genomic data.

Background. The red junglefowl, the wild progenitor of domestic chickens, has historically served as a reference for genomic studies of domestic chickens. These studies have provided insight into the etiology of traits of commercial importance. However, the use of a single reference genome does not capture diversity present among modern breeds, many of which have accumulated molecular changes due to drift and selection. While reference-based resequencing is well-suited to cataloging simple variants such as single nucleotide changes and short insertions and deletions, it is mostly inadequate to discover more complex structural variation in the genome. Results. We present a pangenome for the domestic chicken consisting of thirty assemblies of chickens from different breeds and research lines. We demonstrate how this pangenome can be used to catalog structural variants present in modern breeds and untangle complex nested variation. We show that alignment of short reads from 100 diverse wild and domestic chickens to this pangenome reduces reference bias by 38%, which affects downstream genotyping results. This approach also allows for the accurate genotyping of a large and complex pair of structural variants at the K ‘feathering’ locus using short reads, which would not be possible using a linear reference. Conclusions. We expect that this new paradigm of genomic reference will allow better pinpointing of exact mutations responsible for specific phenotypes, which will in turn be necessary for breeding chickens that meet new sustainability criteria and are resilient to quickly evolving pathogen threats.

Understanding the development of functional attributes of host-associated microbial communities is essential for developing novel microbe-based solutions for sustainable animal production. We applied multi-omics to 388 broiler chicken caecal samples to characterise and model the functional dynamics of 822 bacterial strains. Although microbial community diversity metrics increased with chicken age as expected, the overall metabolic capacity and activity of the microbiota exhibited an unexpected decrease. This drop occurred due to the spread of non-culturable clades with small genomes and low metabolic capacities, including RF39, RF32, and UBA1242. The intensity of this decrease was associated with animal growth, whereby chickens with higher abundances of low-capacity bacteria exhibited higher body weights. This previously unreported link between metabolic capacity of microbes and animal body weight suggests a relevant role of non-culturable bacteria with reduced-genomes for host biology, and opens new avenues in the search for microbe-based solutions to improve sustainability of animal production.

Metagenomic data sets of host-associated microbial communities often contain host DNA that is usually discarded because the amount of data is too low for accurate host genetic analyses. However, if a reference panel is available, genotype imputation can be employed to reconstruct host genotypes and maximise the use of such a priori useless data. We tested the performance of a two-step strategy to input genotypes from four types of reference panels, comprised of deeply sequenced chickens to low-depth host genome (~2x coverage) data recovered from metagenomic samples of chicken intestines. The target chicken population was formed by two broiler breeds and the four reference panels employed were ( i ) an internal panel formed by population-specific individuals, ( ii ) an external panel created from a public database, ( iii ) a combined panel of the previous two, and ( iv ) a diverse panel including more distant populations. Imputation accuracy was high for all tested panels (concordance >0.90), although samples with coverage under 0.28x consistently showed the lowest accuracies. The best imputation performance was achieved by the combined panel due to the high number of imputed variants, including low-frequency ones. However, common population genetics parameters measured to characterise the chicken populations, including observed heterozygosity, nucleotide diversity, pairwise distances and kinship, were only minimally affected by panel choice, with all four panels yielding suitable results for host population characterization and comparison. Likewise, genome scans between the two studied broiler breeds using imputed data with each panel consistently identified the same sweep regions. In conclusion, we show that the applied imputation strategy enables leveraging insofar discarded host DNA to get insights into the genetic structure of host populations, and in doing so, facilitate the implementation of hologenomic approaches that jointly analyse host genomic and microbial metagenomic data.