Genome-Wide SNP Analysis Reveals the Population Structure and the Conservation Status of 23 Italian Chicken Breeds

Abstract

The genomic variability of local Italian chicken breeds, which were monitored under a conservation plan, was studied using single nucleotide polymorphisms (SNPs) to understand their genetic diversity and population structure. A total of 582 samples from 23 local breeds and four commercial stocks were genotyped using the Affymetrix 600 K Chicken SNP Array. In general, the levels of genetic diversity, investigated through different approaches, were lowest in the local chicken breeds compared to those in the commercial stocks. The level of genomic inbreeding, based on runs of homozygosity (F_ROH), was markedly different among the breeds and ranged from 0.121 (Valdarnese) to 0.607 (Siciliana). In all breeds, short runs of homozygosity (ROH) (<4 Mb in length) were more frequent than long segments. The patterns of genetic differentiation, model-based clustering, and neighbor networks showed that most breeds formed non-overlapping clusters and were clearly separate populations, which indicated the presence of gene flow, especially among breeds that originated from the same geographical area. Four genomic regions were identified as hotspots of autozygosity (islands) among the breeds, where the candidate genes are involved in morphological traits, such as body weight and feed conversion ratio. We conclude that the investigated breeds have conserved authentic genetic patterns, and these results can improve conservation strategies; moreover, the conservation of local breeds may play an important role in the local economy as a source of high-quality products for consumers.

Keywords: biodiversity; genetic diversity; local breeds; poultry; safeguard.

Figure 1

Schematic representation of chicken breed origins in Italy (figure taken from http://www.d-maps.com and adapted for illustrative purposes only). For a full definition of breeds, see Table 1.

Figure 2

Genetic relationships among the 27 chicken breeds in this study as inferred by multidimensional scaling (MDS) analysis using (A) the breed-average coordinates of eigenvalues of C1 and C2 and (B) all of the individuals per breed. Breed acronyms are reported in Table 1.

Figure 3

Maximum likelihood estimation calculated with the admixture algorithm. The inferred clusters (K) were represented from K = 2 to 27. Breed acronyms are reported in Table 1.

Figure 4

Neighbor-joining tree constructed on the Reynold’s genetic distance for the breeds considered (A) and based on individual allele-sharing distances (B). Breed acronyms are reported in Table 1.

Figure 5

Boxplot of the inbreeding coefficient (F_ROH) estimated from runs of homozygosity for each breed considered in this study.

Figure 6

Manhattan plot of each single nucleotide polymorphism (SNP) significance in runs of homozygosity. Blue line indicates the top 0.999% of SNPs.