Estimation of genetic and demographic parameters of extensively raised chicken populations using genome-wide single nucleotides polymorphism (SNP) data.

Abstract

Village chicken populations are raised under a farming system that faces a number of challenges such as small flock size, lack of animal performance and pedigree records, lack of proper husbandry and poorly defined and structured mating systems all of which can negatively influence the genetic structure of the populations. Understanding of the evolutionary history, demographic structure, inbreeding levels and risk of a population to extinction is important in facilitating genetic improvement programs while maintaining biodiversity of extensively raised chicken populations. Linkage disequilibrium (LD) is an important source of information about historical events of recombination in a population and together with an understanding of the haplotype structure can provide valuable guidelines for breed improvement. This study was undertaken to investigate the existing LD level, inbreeding levels, effective population size and haplotype structure of extensively raised chicken populations from Southern Africa. A total of 312 village chickens from Malawi (n = 30, from one ecotype), South Africa, (n = 146, from three different ecotypes) and Zimbabwe (n = 135, from three different ecotypes) were genotyped using the Illumina chicken iSelect SNP60K bead chip. LD was calculated for each population from a total of 43,175 SNP after pruning for minor allele frequency (MAF) <0.05, genotyping call rate of <0.95, and deviation from Hardy Weinberg Equilibrium (HWE) p <0.001 and missing genotypes of more than 5%. Linkage disequilibrium averaged 0.41±0.006 and was observed to extend up to a marker distance of 100 kb. From the LD, effective population size was estimated that indicated reduced size of the breeding population over the past 40 generations to less than 20 individuals. Haplo-block structure analysis resulted in a total of 649, 2104 and 2442 blocks from Malawi, South Africa and Zimbabwe, respectively. Most of the observed blocks were less than 20 kb with a few that were more than 500 kb. Haplo-block genome coverage was 39 Mbp, 64.4 Mbp and 54.5 Mbp for Malawi, South Africa, and Zimbabwe, respectively. Large haplo-blocks on chromosome 8 spanned QTL regions associated mostly with body composition traits. The LD pattern was consistent with low effective population sizes and loss of heterozygosity in the village chicken populations. Potentially useful haplo-blocks spanning regions of known QTLs should be targeted for further analysis and identification of genes conferring optimal production performance of village chickens under harsh and marginalized production systems. Overall, the study provides baseline information on the utility of genome wide SNP data in studying extensively raised village chicken populations.