Effective population sizes in cattle, sheep, horses, pigs and goats estimated from census and herdbook data

Abstract

Accurate measures of effective population sizes ( N e ) in livestock require good quality data and specialized skills for their computation and analysis. N e can be estimated by Wright's equation N e = 4MF/( M + F) (M, F being sires and dams, respectively), but this requires assumptions which are often not met. Total census sizes N c of livestock breeds are collated globally. This paper investigates whether estimates of N e can be made from N c ; this would facilitate conservation monitoring. Some N e methodologies avoid the assumptions of Wright's equation and permit measurement, rather than estimation, of N e . Those considered here employ, respectively, linkage disequilibrium (LD) of single-nucleotide polymorphisms (yielding N e ( LD)), and genealogical analysis (rate of increase of inbreeding, DF), yielding N e (DF). Considering breeds of cattle, sheep, horses, pigs and goats for which N c and either N e ( LD) or N e ( DF) are known (totals of 203 breeds and 321 breeds, respectively), proportionality has been investigated between N c and these measures of N e . N e ( LD) was found to increase with N c , significantly in sheep and horses, less so in cattle, but not at all in pigs. N e ( DF) was correlated with log 10 ( N c ) in cattle, sheep and horses (53, 56, 43 breeds, respectively). N e ( LD) was correlated in cattle (73 breeds) and pigs (31 breeds) with the log 10 transformation of N e as calculated by Wright's equation.Further verification and refinement are needed, particularly of census data, but credible predictions of N e are obtainable by applying the following multipliers to log 10 ( N c ): cattle 17.61, sheep 97.72, horse 70.78. For cattle and pigs, multiplying log 10 ( N e (Wright)) by, respectively, 40.69 and 60.09, also gives credible predictions. Such census-based estimates of N e could in principle be generated by non-specialists and are likely to be suited to audits of conservation activity when financial resources or availability of data are limiting. The ratio N e /N c varied among species with an overall median value of 0.03, less than a tenth of that typically observed in wild mammals. Characteristics were also investigated of a distinct herdbook-based methodology, namely the development of Wright's equation to take into account variances of progeny numbers to yield what has been termed here N e (Hill). Comparison of these values with N e (Wright) could help to identify breeds with breeding structures conducive or inimical to genetic conservation. However, N e (Hill) requires breed-specific values for these variances, and this restricts its applicability.

Bovine breed-specific augmented reference graphs facilitate accurate sequence read mapping and unbiased variant discovery

“…Considering that most cattle breeds have an effective population size between 50 and 200 [38,39], the vast majority of variants with allele frequency greater than 0.1 can be detected from a few sequenced key ancestor animals [13]. Most of the BSW, FV, OBV and HOL animals of our study are also key ancestor animals [7,12,13,40].…”

Benefits of Dominance over Additive Models for the Estimation of Average Effects in the Presence of Dominance

“…We purposefully used relatively small sample sizes to illustrate the effect. Although sample sizes taken in empirical studies may be larger, effective sample size may be much smaller, because actual populations often have small effective population size ($N_e$) (Hall 2016). This low $N_e$ is related to the family structure in the population, where many individuals are bred from a limited number of parents, so that $N_e$$\text{≪ }\mathrm{normalN}$.…”

Genomic characterisation and conservation genetics of the indigenous Irish Kerry cattle breed

“…From a conservation perspective, livestock populations generally exhibit N e values relative to total census population sizes ( N c ) that are substantially lower than seen in comparable wild mammal populations (Hall, 2016). Also, estimation of N e using methods such as SNeP that leverage genome-wide SNP linkage disequilibrium (LD) data will tend to underestimate N e because of physical linkage between many of the SNPs in the data set (Waples et al, 2016).…”