The ChickenGTEx pilot analysis: a reference of regulatory variants across 28 chicken tissues

Abstract

Chicken is a valuable model for understanding fundamental biology, vertebrate evolution and diseases, as well as a major source of nutrient-dense and lean-protein-enriched food globally. Although it is the first non-mammalian amniote genome to be sequenced, the chicken genome still lacks a systematic characterization of functional impacts of genetic variants. Here, through integrating 7,015 RNA-Seq and 2,869 whole-genome sequence data, the Chicken Genotype-Tissue Expression (ChickenGTEx) project presents the pilot reference of regulatory variants in 28 chicken tissue transcriptomes, including millions of regulatory effects on primary expression (including protein-coding genes, lncRNA and exon) and post-transcriptional modifications (alternative splicing and 3 untranslated region alternative polyadenylation). We explored the tissue-sharing and context-specificity of these regulatory variants, their underlying molecular mechanisms of action, and their utility in interpreting adaptation and genome-wide associations of 108 chicken complex traits. Finally, we illustrated shared and lineage-specific features of gene regulation between chickens and mammals, and demonstrated how the ChickenGTEx resource can further assist with translating genetic findings across species.

RNA sequencing variants are enriched for eQTL in cattle tissues

“…However, a strong depletion of non-coding variants in typical RNA-seq datasets results in a less reliable imputation of variants that are distant to transcribed regions. Similar observations have been made in chicken (11), pig (12), and human (13).…”

“…Livestock GTEx consortia rely on RNA-seq for variant calling (e.g., cattle (8), pig (12), and chicken (11)) to enable molecular QTL mapping as most RNA-sequenced samples don’t have matched DNA-based genotypes or sequences. This is different to the equivalent human GTEx (13) which uses transcriptomes that have matched DNA whole-genome sequencing.…”

“…Furthermore, the low agreement we observed for the top associated variants between DNA-seq and RNA-seq variants would weaken downstream analyses like colocalization of putative causal variants (4) if depending only on RNA-seq variants. Livestock GTEx consortia rely on RNA-seq for variant calling (e.g., cattle (8), pig (12), and chicken (11)) to enable molecular QTL mapping as most RNA-sequenced samples don't have matched DNA-based genotypes or sequences. This is different to the equivalent human GTEx (13) which uses transcriptomes that have matched DNA whole-genome sequencing.…”

Integrated Global Chicken Reference Panel From 13,187 Chicken Genomes

“…This extensive collection will serve as a valuable genetic database for studying the evolution and improvement of chickens and other birds. Additionally, our project will collaborate with the FAANG project (43) and gradually integrate the GCRP with regulatory elements atlas (44), and ChickenGTEx (45). This integration can then be applied to meta-and colocalization analyses (46,47), facilitating a more comprehensive analysis of gene regulatory mechanisms.…”

Empowering bioinformatics communities with Nextflow and nf-core

“…For instance, AQUA-FAANG relied on nf-core pipelines to exploit the evolutionary conservation of genomic regulatory elements and epigenetic states across six farmed fish species for multiple biological conditions and matched tissue panels 29 . Data integration across projects is critical, as projects such as the farm animal GTEx [30][31][32] are expected to keep integrating a larger number of species and deeper sequencing over time. Interoperability, achieved through the use of a common analytic framework is therefore an essential component of long-term sustainability, well beyond the goal and lifetime of projects like EuroFAANG.…”