Local cattle breeds continue to decline in numbers partly due to the use of high performing breeds in advanced production systems where genetic material of elite animals is widely spread. The objective of this study was to assess the within and across breed genetic diversity of the Angler and Red-and-White dual-purpose (DP) cattle breeds applying different inbreeding concepts. Classical and ancestral inbreeding coefficients were computed from pedigree data using the gene dropping method. Effective population size was calculated based on the increase of classical inbreeding, and based on ancestral inbreeding to obtain what was termed as ancestral effective population size. Furthermore, the effective number of founders and ancestors were computed to assess the disequilibrium of founder contribution in the reference populations. The analyses were performed separately for each breed and for a combined dataset. The Angler pedigree was more complete (88%) in the first parental generation but completeness declined with increasing pedigree depth. Average classical inbreeding coefficients of inbred individuals were 2.19%, 1.94% and 2.07%, while average Ballou’s ancestral inbreeding coefficients were 3.69%, 1.39% and 2.21% for the Angler, Red-and-White DP and the combined breed pedigree analyses, respectively. Ancestral history coefficient is a novel coefficient and its estimates were similar and strongly correlated to Ballou’s coefficients (r = 0.99, p < 0.001). The effective population size estimates ranged from 156 to 170 for the classical inbreeding based method, and as low as from 50 to 54 for the ancestral history coefficient based method. The effective number of founders and ancestors ranged from 310 to 532, and 90 to 189, respectively. Our results show that the Red Holstein breed is a key progenitor of the breed populations under study. This highlights cross breeding schemes introduced to improve the milk trait performance of the Angler and Red-and-White DP breeds some decades ago.
Angler (RVA) and Red-and-White DP (RDN) cattle are local breeds of German origin. Both breeds have small populations predominantly found in the Northern part of the country. Planned breeding of RVA dates back to 1838, however, the organisation of a central herdbook and official milk recording began in 1879 and 1902, respectively [
Available data from the German Federal Office for Agriculture and Food indicate a gradual decline in herdbook number of RVA and RDN bulls and cows over the past two decades [
In the current study, we performed both within and across breed diversity assessment by calculating classical inbreeding coefficients, ancestral inbreeding coefficients according to Ballou [
Pedigree data for RVA and RDN span the period between 1906 and 2016 and were obtained from the official computation centre responsible for breeding value estimation (Vereinigte Informationssysteme Tierhaltung w. V., Verden, Germany). The RVA dataset consisted of 93,078 animals, including 10,481 bulls and 82,597 cows. A total of 184,358 animals including 16,068 bulls and 168,290 cows formed the RDN pedigree dataset. For both breeds, there has been some form of introgression of genetic material from other conventional breeds including the German Black-and-White Holstein (SBT), German Red-and-White Holstein (RBT), Holstein from North America (HOL), Jersey (JER), Braunvieh (BV), Fleckvieh (FV) and Scandinavian Red cattle. Breeds other than RVA and RDN were therefore present in both datasets, and consequently, 12,709 animals were common to both pedigrees. The completeness of pedigree [
Using SAS software (SAS 9.4, SAS Institute Cary, NC, USA), we recoded the animal identification numbers (Ids) in the raw pedigree file from 15 digits to 14 digits, the maximum number required by PEDIG software [
The classical individual inbreeding coefficient (F), defined as the probability of an individual having two identical alleles by descent, was calculated following [
Defined as the number of individuals in an ideal population that would give rise to the same rate of inbreeding as observed in the actual breeding population [
N e = 1 / 2 b (1)
Additionally, we applied the same estimation procedure in the calculation of what we termed as ancestral effective population size, which can be defined as the size of a population as reflected by its rate of ancestral inbreeding. In this regard, three values being Ne_Bal, Ne_Kal and Ne_AHC were distinguishable.
The effective number of founders ( f e ) and effective number of ancestors ( f a ) best describe the unbalanced representation of founder contributions in a reference population. Parameter f e defines the number of equally contributing founders that would be expected to produce the same genetic diversity as in the population under study [
f e = 1 / ∑ k = 1 f q k 2 (2)
where q k is the probability that a gene randomly sampled in the population originates from founder k, and f is the total number of founders [
f a = 1 / ∑ j = 1 a q j 2 (3)
In Equation (3), q j represents the marginal genetic contribution of ancestor j, i.e. the genetic contribution made by an ancestor that is not explained by previously chosen ancestors, and a is the total number of ancestors considered. To calculate marginal genetic contributions, the first major ancestor was found based on its raw genetic contribution (i.e. qk = qj) following an iterative procedure. Next, the genetic contribution of the nth major ancestor was calculated conditional on the genetic contribution of the n − 1 already chosen ancestors. Reference [
3.73 for RVA, RDN and RVA_RDN, respectively, and consistent with the ranking of the three scenarios based on the trends in pedigree completeness across parental generations. The equivalent complete generation is an appropriate criterion to characterise pedigrees [
Similar to the results of previous studies, there is a general trend of decreasing pedigree completeness with increasing pedigree depth. Pedigree recording in the study populations started over a century ago and at a time when little was known about planned breeding. Recognition of breed importance and improvements achieved in breeding over the years are contributing factors to the observed increase in data recording from founder to recent generations. Incompleteness of pedigrees in this study implies a caution about the overreliance on our data for inbreeding estimation. It was demonstrated that with only 10% of unknown pedigree, inbreeding is strongly underestimated [
The numbers of inbred individuals were 59,000, 39,477 and 95,343 representing 64%, 21% and 36% for RVA, RDN and RVA_RDN, respectively. The percentage of inbred individuals was low for the RDN pedigree and this is due to the inability of the pedigree data to fully capture the relationships between all animal as discussed previously.
Item | RVA (%) | RDN (%) | RVA_RDN (%) |
---|---|---|---|
F for all animals | 1.39 | 0.41 | 0.75 |
F for inbred animals | 2.19 | 1.94 | 2.07 |
F_Bal | 3.69 | 1.39 | 2.21 |
F_Kal | 0.16 | 0.05 | 0.09 |
AHC | 3.94 | 1.49 | 2.37 |
(3.25%) [
Knowing the population level inbreeding rate alone is not enough, rather, the effect of inbreeding as manifested in the reduction in individual’s performance per unit increase in inbreeding coefficient (i.e. inbreeding depression). Ballou’s concept of ancestral inbreeding proposes a measure that tells which individuals or population harbour fewer detrimental genes. Thus, higher values of the parameter indicate the likelihood of an individual having fewer detrimental genes. Following this concept, it can merely be said that the RVA breed population has endured high inbreeding at the ancestral level (F_Bal = 3.69%) and is probably prone to fewer incidents of inbreeding depression. The mean estimates for F_Kal were much lower, i.e. 0.16%, 0.05% and 0.09% for RVA, RDN and RVA_RDN, respectively. By definition, F_Kal deals with alleles which are homozygous because they have met in the past, and only includes the ancestral inbreeding of relationship. This means that unlike F_Bal, F_Kal for an individual remains zero when its classical inbreeding coefficient is zero. Note, that our analysis did not include the second component of the parameter that deals with new inbreeding. To our knowledge, the results on AHC in this study represent one of the first tests of this coefficient using real data. Estimates of AHC were high i.e. 3.94%, 1.49% and 2.37% for RVA, RDN and RVA_RDN, respectively, and very similar to the estimates of F_Bal. The advantage of AHC is that it offers an appropriate measure of inbreeding when selection against deleterious recessive alleles is less than fully efficient. The correlation between the different inbreeding coefficients are
presented in
Estimates of effective population size are given in
Parameter | F_Meuw | F_Gendrop | Fa_Bal | AHC | Fa_Kal |
---|---|---|---|---|---|
F_Meuw | - | 0.999 | 0.506 | 0.502 | 0.748 |
F_Gendrop | 0.999 | - | 0.506 | 0.502 | 0.748 |
Fa_Bal | 0.264 | 0.264 | - | 0.998 | 0.658 |
AHC | 0.266 | 0.266 | 0.997 | - | 0.669 |
Fa_Kal | 0.679 | 0.679 | 0.634 | 0.648 | - |
Parameter | RVA | RDN | RVA_RDN |
---|---|---|---|
Ne | 156 | 170 | 161 |
Ne_Bal | 54 | 59 | 58 |
Ne_Kal | 1040 | 1186 | 1160 |
Ne_AHC | 50 | 54 | 53 |
the lowest average inbreeding coefficient but also the poorest pedigree quality. Here, effective population size values were estimated by regressing the individual inbreeding coefficients on the equivalent complete generations traced and considering the regression coefficient as the rate of inbreeding. The same procedure was applied to the individual estimates of Fa_Bal, Fa_Kal and AHC to calculate for the first time ancestral effective population size which we defined as the size of a population as reflected by its rate of ancestral inbreeding. The ancestral effective population size estimates based on Fa_Bal (Ne_Bal) and that based on AHC (Ne_AHC) were similar and ranged from 50 to 59 animals for all data considerations. Estimates of ancestral effective population size based on Fa_Kal (Ne_Kal) on the other hand, were unrealistically high and above 1000 animals. Nevertheless, these high estimates are not surprising since Fa_Kal considers only “old” inbreeding. Applying different computation methods [
The parameters derived from the probability of gene origin account for the unbalanced use of founders in a pedigree and unlike N e , are less affected by pedigree errors [
Item | RVA | RDN | RVA_RDN |
---|---|---|---|
Total number of animals (N) | 93,078 | 184,358 | 264,727 |
Animals with both parents known (RP) | 76,520 | 73,749 | 142,240 |
Base population (N − RP) | 16,558 | 110,609 | 122,487 |
Ancestors contributing to reference population | 10,059 | 24,101 | 30,911 |
Effective number of founders | 310 | 519 | 532 |
Effective number of ancestors | 90 | 189 | 159 |
fa/fe | 0.29 | 0.36 | 0.30 |
pedigree had a slightly lower RP number (73,749) although it has the highest total number of animals. A total of 10,059, 24,101 and 30,911 ancestors, some of which were not founders contributed to the RVA, RDN and RVA_RDN reference populations, respectively.
The f e values obtained were 310 (RVA), 519 (RDN) and 532 (RVA_RDN). Published f e values for other cattle breeds range from 40 to 649 animals [
The marginal genetic contribution of the top 10 ancestors to the RVA, RDN and RVA_RDN reference populations are given in
Ancestor ID | Sex | Birth Year | Breed Type | Marginal contribution | Offspring |
---|---|---|---|---|---|
RVA | |||||
840000001842371 | Male | 1980 | Red & White Holstein | 0.045423 | 167 |
840000001629391a | Male | 1972 | Red & White Holstein | 0.029506 | 135 |
276000102168990d | Male | 1974 | Angler | 0.027598 | 244 |
276002240018965e | Male | 1966 | Angler | 0.027072 | 328 |
840000001491007b | Male | 1965 | Red & White Holstein | 0.025572 | 231 |
752000000093907f | Male | 1990 | Angler | 0.024679 | 184 |
840000001427381c | Male | 1962 | Red & White Holstein | 0.023268 | 93 |
000008400028756 | Male | 1963 | Angler | 0.021919 | 19 |
528000775157228 | Male | 1991 | Red & White Holstein | 0.019265 | 147 |
276000102142217 | Male | 1970 | Angler | 0.018949 | 605 |
RDN | |||||
840000001629391a | Male | 1972 | Red & White Holstein | 0.031539 | 245 |
840000001491007b | Male | 1965 | Red & White Holstein | 0.025185 | 323 |
528000000355040 | Male | 1973 | Red & White (RDN) | 0.018695 | 166 |
124000000267150g | Male | 1958 | Red & White Holstein | 0.018484 | 59 |
528000000338535 | Male | 1971 | Red & White Holstein | 0.017550 | 47 |
840000001427381c | Male | 1962 | Red & White Holstein | 0.017217 | 94 |
840000001189870 | Male | 1952 | Red & White Holstein | 0.014265 | 90 |
000009002053500 | Male | 1966 | Red & White Holstein | 0.013015 | 349 |
528000951276374 | Male | 1982 | Red & White (RDN) | 0.011961 | 25 |
000009002037187 | Male | 1965 | Red & White Holstein | 0.011538 | 116 |
RVA_RDN | |||||
840000001629391a | Male | 1972 | Red & White Holstein | 0.037133 | 279 |
840000001620273 | Male | 1972 | Red & White Holstein | 0.025723 | 24 |
840000001491007b | Male | 1965 | Red & White Holstein | 0.024239 | 377 |
840000001427381c | Male | 1962 | Red & White Holstein | 0.019417 | 114 |
276000102168990d | Male | 1974 | Angler | 0.014904 | 245 |
276002240018965e | Male | 1966 | Angler | 0.014614 | 329 |
124000000267150g | Male | 1958 | Red & White Holstein | 0.014474 | 61 |
752000000093907f | Male | 1990 | Angler | 0.013290 | 184 |
840000000005304 | Female | 1973 | Red & White Holstein | 0.012795 | 2 |
840000001189870 | Male | 1952 | Red & White Holstein | 0.012364 | 92 |
a-gAncestor IDs with the same superscript indicate the same animal appearing in the different pedigree datasets (Ancestors were selected based on marginal contribution calculated following [
tion of Angler and Red-and-White DP cattle breeds are genetically not distinct. In fact, they share common ancestors some of which can be traced back to as early as 1965. Most striking is the high genetic contribution of the Red-and-White Holstein breed to the breed populations under study. For the Red-and-White dual purpose breed, it has been established that a pedigree of the breed has a maximum of 25% Red Holstein genes [
Analysing the Angler and Red-and-White dual-purpose local cattle pedigrees has shed some light on the population structure of these breeds in Germany. The current study demonstrates that Ballou’s approach to estimate ancestral inbreeding and the novel ancestral history coefficients are similar approaches that produce comparable results. Besides, these coefficients provide avenue to calculate effective population size at the ancestral level. The effective population size of the breeds did not raise concern, however, due to incompleteness of the pedigree data used, consideration of the parameters derived from the probability of gene origin was extremely necessary in characterising the genetic diversity within the populations. For both breeds, the reference populations were raised from founder or ancestor groups, within which genetic contributions were typically unbalanced, male animals being favoured. Consequently, only a few animals explained the complete genetic diversity in the population under study. The Red Holstein breed is a key progenitor of the current Angler and Red-and-White dual-purpose cattle populations in Germany. Based on the high genetic contribution of key ancestors belonging to other breeds, we recommend an extensive investigation of foreign blood percentage in both breeds.
This work was conducted as part of the research activities of the operational group, “Animal Genetic Resources” that operates under the auspices of the Agricultural European Innovation Partnership (EIP-AGRI) project. The authors are thankful to the European Commission for providing funds for the project. Personnel at the “Vereinigte Informationssysteme Tierhaltung” in Lower Saxony are also acknowledged for the provision of data and their relentless efforts in answering questions about the datasets used.
Addo, S., Schäler, J., Hinrichs, D. and Thaller, G. (2017) Genetic Diversity and Ancestral History of the German Angler and the Red-and-White Dual-Purpose Cattle Breeds Assessed through Pedigree Analysis. Agricultural Sciences, 8, 1033-1047. https://doi.org/10.4236/as.2017.89075