Jump to content
Search In
  • More options...
Find results that contain...
Find results in...

Welcome to American Indian Dogs

Register or login now and gain instant access to our features!

Pedigree Analysis Of The Aidog


Recommended Posts

Review by Kim La Flamme:

 

Peter Jesenak (AI dog owner) with partners Dr. Erik Postma, Prof. Dr. Lukas F. Keller, at Zürich University and Dr. Paul Jesenak, ETH Zürich, plus Monica Santana, dipl. designer HGKZ; all have done a great job on these new modern research / analysis papers. This will help a lot for our future AI dog registry, enabling us to instantly look up any facts within these papers that we bring in from our AI dog registry or individual dogs. Even more so, once we add the many past generations, beyond these six generations contained here.

 

 

 

For an example we can type in how many blue eyed dogs compared to yellow, genetic color recessives and dilutes and % of these colors that could be lost, most used stud dogs and dams in the modern history of the breed, or any health problems that might be inherited ...etc, with endless uses and possibilities. Modern science and logical methods to better “save” ancient animals, or for that matter ourselves.

 

 

 

What really stands out is when you put a typical AKC type breed through this same analysis. It's amazing the difference of inbreeding within the graphs that shows up. Instead of a few lines coming together to a few individual breeding dogs spread out through just a few generations; as within the AI dog breed or almost extinct wild animal breeding programs. All of the AKC breeds shows the entire breed being developed from just a very few favorite "show conformation champion" sires and dams for many, many generations, with no out crossings brought in for 100's of years, with so many “potential” healthy variables lost, forever.

 

 

 

It's certainly no wonder modern “breeds” are so unhealthy, both in the mind and body. This shows right there on the graphs that all AKC breeds come from just a very few dogs; this means more health problems coming from 3 or 4 dogs that essentially make up the entire breed. With so many criss-crossed lines leading to just a few dogs, it blurs out the entire page.

 

 

 

Registering breeding graphs like this might help animal breeders understand a better way to save both domestic and wild animals from becoming instinct and encouraging healthier breeding practices for the health of animals in the future... and of coarse there will always be favorite sires and dams .. even in nature with survival of the fittest (form follows function) it works that way naturally. But nowhere near the same type of "show quality" type selection that happens in our modern times, with all animals.

 

 

 

Thanks to Peter Jesenak and his team at Zürich University.... great new way to be able to easily look at an entire animal breeding program and how it is being bred and cared for ... or not.

 

 

 

 

 

Kim La Flamme: founder/trustee of the AI dog

 

________________________________________________________________________________

_________________________________________________________

 

 

Pedigree analysis of the American Indian Dog

 

Julian Amacker, Peter Jesenak

Block course Biology, Bio 364 Principles of Evolution, University of Zürich

3.November 2008 – 19. December 2008

 

 

__________________________________________

 

ABSTRACT: Conservation genetics is becoming more and more important to preserve a healthy diversity of species. Problems occur especially when populations are small and the breeding is not random. The project described in this paper analyses the pedigree of the American Indian Dog, an old breed from North America that was close to extinction nearly 50 years ago but was saved and is meanwhile recovering. Starting out with a small population of founders, inbreeding and close kinship appear to be the problem as the goal is to keep the breed healthy and with certain genetic variation.

Unlike most other breeds, the studbook of this breed remains open, which allows the introduction of new genes over time and therefore maintaining the diversity. We analyzed how inbreeding coefficients are passed on through time and how they become influenced by newly introduced animals. The data describes how the inclusion of unrelated individuals

into an existing population takes effect on its genetic variation. It also shows what happens when one individual becomes overbred and the consequences for the relatedness of future generations, i.e. increasing inbreeding and loss of heterozygosity. Also, we compared the current breeding strategy with a random mating population and concluded that when breeding dogs, certain preferences (conscious or unconscious) such as the color or character properties occur that lead to a higher inbreeding coefficient and a certain loss of heterozygosity compared to a random mating population. Nevertheless, besides keeping the inbreeding coefficients low, the main goal of breeding remains to maintain a breed with physically and mentally healthy individuals.

Key Words: American Indian Dog, Pedigree, Inbreeding, Kinship, Heterozygosity, Generation Number, Random Mating.

 

_____________________________________________________________________

INTRODUCTION

 

In the scope of the block course “Principles of Evolution” at the University of Zurich, it was our task to find a topic and process it in the time frame of seven weeks. Our choice

was to conduct a pedigree analysis on the American Indian

Dog. The reason we choose this topic was our interest in conservation genetics, especially in conjunction with this special breed. The breed can be considered rare, since there are only about 600 individuals left around the world. The fact that the population size is relatively small leads to certain difficulties considering the maintaining of this breed. Mating under relatives, called inbreeding, is almost impossible to avoid and therefore the probability of having two alleles at a locus that are identical by descent (inbreeding coefficient F) adds up through time which results in more recessive deleterious traits manifesting themselves. This effect even gets amplified by popular sires (males with higher offspring number than average, Figure 1) and can eventually lead to an inbreeding depression, i.e. a reduced fitness.

 

The American Indian Dog differs in a number of points from other breeds. It probably belongs to the oldest domesticated dogs, going back for over 10,000 years, when the Native Americans began to use dogs for several purposes.

 

Back then, the dogs were involved in hunting, pulling sleds and travois, guarding the tribe or herding to name just a few of the diverse tasks from those past times. Today,

this breed is used in a more modern way for agility, Frisbee dogs, search and rescue or as family pets just to give an example.

The breeding strategy in the times of the Indians, according

to Kim LaFlammes research, was very different compared to how dogs are being bred today. Dogs living in different tribes were brought to gatherings and were traded

among the different groups. No breed standards (rules

defining the dogs appearance) existed back then and the genetic diversity could be maintained at a high level. Selection of the breeders occurred according to the dog’s abilities and accomplishments and not in the first point to its appearance or look. Close inbreeding was avoided.

Today most dogs are being bred according to a breed standard and the breeding dogs are being chosen by their success at shows. This approach leads to close inbreeding followed by physical and to some extent also to mental problems, as can be seen among modern breeds like the German Sheppard or Labrador Retriever, Maki K. et al. (2001). The whole process is being stimulated by the big kennel clubs and dog registries, which reinforce the compliance of breeding standards and do not allow new dogs and therefore new genes in their studbooks. The American Indian Dog is not registered with AKC (American Kinology Club) or UKC (United Kennel Club) but with an independent

registry, the IPDBA (International Progressive Dog Breeders Alliance) which leaves the breeder in full control of the breed without constraints.

 

About 50 years ago, the native dogs of America were considered to be almost extinct, with just a few remaining at the widely spread tribes. Kim LaFlamme took the quest to save these special companions and enable them to survive in modern society. He traveled trough the country and gathered different dogs from distinct tribes to establish a foundation to start breeding and preserving the dogs. His philosophy was to keep them just the way they used to be in the old times according to his research and to keep on breeding practice of the Native Americans. The studbook remains open which gives the possibility to take in new dogs and therefore keep up the genetic variation and health

of the breed, Bouzat J. et al. (1998) & Westemeier et al. (1998).

 

At the main kennel in Oregon, USA, there are in average about 25 breeding dogs present, continually changing according to new matings. The breeders’ ambition is to keep a divergence in the breed, concerning color variation, body shape, character, abilities and overall look. This intention can be difficult to unite with the notion of avoiding inbreeding. Although inbreeding appears universally to reduce the fitness, its magnitude in specific effects is highly variable because they depend on the genetic constitution of species or populations, Hedrick & Kalinowski 2000. This means that selection for desired traits can be accelerated by inbreeding, if advantageous alleles get fixed.

 

The goal of the present project was to analyze the degree of inbreeding and its interrelationship with introducing unrelated dogs or ferals. Does the practiced introduction of new genes reduce the inbreeding and maintain genetic variation? How do the color patterns correlate with inbreeding? Is there an effort to avoid a loss of heterozygosity?

 

__________________________________________

 

METHODS

 

To start the project, as much data as possible had to be collected, concerning mating, color, date of birth and so on. This appeared to be more difficult then expected. Finally

we received the necessary information from the IPDBA, containing data from 2003 – 2008. Data from before 2003 was not available at the registry and would have had to be copied by the breeder and submitted by mail, which was not possible in the given time period. Finally, after three weeks of negotiations, the data was made available and the project could start.

 

First, the data from the IPDBA (in *.bxf format) had to be translated into a practical Excel-sheet. The raw material was sorted in clusters, which had to be split up into single columns in Excel. We mostly used the “VLookup” function, to extract and sort the data for all further analyses. These clusters included 208 individuals with their registration number, parents, sex, date of birth and color. Although the data of the last two traits was not complete (63% available), we used the present color data to analyze if there is a dominance in this trait. Further a four-column sheet (txt-format), with ID’s, sires, dams and effective generation number (GnE) had to be created to import the data into Pedigree Viewer, a software suitable to compute inbreeding coefficients and to construct the pedigree, Kinghorn B & S (2005).

 

According to the information of Kim LaFlamme the American Indian Dog is being officially bred since approximately 50 years, but the registration with IPDBA had started only in 2003. Due to this fact there are 45 individuals without a registered sire and dam and one individual without a registered dam in the present data. 33 of those are real founders, meaning that these dogs were newly introduced into the breeding program and are not at all related to the rest of the dogs.

 

Pedigree Viewer was used for a general overview of the pedigree (Figure 1). Further we calculated the inbreeding coefficients (F) for all individuals also using Pedigree

Viewer. The inbreeding coefficient is defined as the probability that two alleles are identical by descent. We calculated the average inbreeding coefficient for each generation

to show tendencies of inbreeding. To do so, we applied two different ways to assign a generation number to each individual. Generation numbers computed by Pedigree

Viewer (GnP), do not give information about the diversity of biological generations. Pedigreviewer assigns a generation number to the offspring that is one higher than the maximal generation number of the parents, e.g. offspring GnP = 6 if parents have GnP = 1 and GnP = 5. The advantage of structuring the population in this way is having an approximate overview where new founders were introduced through time (Figure 1). Because there are many overlapping generations we used the definition of Brinks et al. (1962) to calculate the “effective generation number” (GnE):

 

equ1.JPG (1)

 

 

Where GnEi is the generation number of an individual with parents of generation number GnES and GnED.

 

This definition allows to calculate the average inbreeding coefficient from the “biological generations” and also the effective population size (see below).

 

Furthermore we defined two different population sizes, one with all the individuals of a generation (PA), another including only breeding animals (PB). In later case we also estimated the effective population size Ne to be:

 

equ2.JPG (2)

 

Where NS is the number of sires and ND the number of dams, Evolutionary Analysis (2007).

 

fig1.JPG

 

 

We then plotted the average inbreeding coefficients against GnP and GnB over time for all three population sizes, resulting in five charts. Besides the mean inbreeding coefficient we also included the mean value for the sire and for the dam respectively.

To figure out how much of the inbreeding is simply due to a small population size where random mating leads to a high value of inbreeding, and how much is due to non-random mating, we calculated the mean kinship coefficient

of all mating individuals in GnE by using SAS. This happened with the help and support of Erik Postma.

 

equ3.JPG (3)

 

Where FA is the inbreeding coefficient of the path’s common ancestor and FS and FD are the inbreeding coefficients of sire and dam. Inbreeding coefficients of the ancestors

are used for this calculation because kinship is defined as the probability that random sampled of alleles out of two homologous alleles of an animal x is identical by descent

to a randomly sampled allele of animal y, Malecot G. (1948).

 

This coefficient equals the inbreeding coefficient of a fictitious offspring generation under random mating. We then shifted each mean inbreeding coefficient (from all dogs of one GnE) one generation back to compare it with the mean kinship of this generation. Any discrepancy between the two values shows the grade of consanguineous mating, Calboli et al. (2008). Furthermore the kinship coefficient shows the maximal avoidance of loss of heterozygosis possible, if the breeding was performed according to random mating.

 

To analyze which factors lead to the calculated inbreeding coefficient we used two methods:

 

1. We compared two different methods to calculate the change in heterozygosity. First method (formulas):

 

equ4.JPG (4)

 

Where ΔH is the change in heterozygosity.

 

This shows the loss of heterozygosity due to the actual inbreeding coefficient. The second method:

 

equ5.JPG(5)

 

shows the loss of heterozygosity due to inbreeding caused by a small effective population size. This formula cannot be used for large population sizes, because large populations

require the inclusion of the previous heterozygosity. If in a small population ΔH1

 

2. We compared the mean inbreeding coefficients of all color-groups, to see if there is a linkage between them. A significantly higher average inbreeding coefficient for a color could show a preference in breeding for this color, and leads therefore to a higher inbreeding coefficient.

 

Finally we evaluated the dominance of the trait “color” by making a pivot-table in Excel, which shows numbers of offspring of a certain color descended from different combinations of parental colors, i.e. if offspring has a color that does not appear in the parents, then this color cannot be dominant.

 

chart1_4.JPG

 

chart5.JPG

 

_____________________________________________________________________

 

 

RESULTS

 

Population size and generation number: The difference in generation size between GnP and GnE are dramatically (chart 1 & 2). Compared to GnP, the GnE shows much higher amount of individuals in generation zero. GnP and GnE are similar in decrease of the number of breeders in the last three generations.

 

Also, the number of generations differs from eleven, counted in Pedigree Viewer considering all dogs, to eight, calculated as effective population size (Ne). Ne is very close to the values of GnE where only the breeding dogs were taken in consideration (chart 3). GnE three, four and six, show relatively big differences between the number of males and females in one generation. These generations can be described as bottlenecks.

 

Inbreeding coefficient: The mean inbreeding coefficient rises from 0 in generation zero to 0.2 in the last generation (chart 2 & 3). Some discrepancy between the values from the male and the female F can be observed. A strong increase after generation six can also be observed.

 

As shown in chart 3 there is a strong discrepancy between the actual inbreeding coefficient (non-random mating) after generation five and the one expected under random mating. In generation eight, the actual value of the inbreeding coefficient is 0.23, compared to 0.03, which could be the lowest possible coefficient under random mating. This fact is supported by our analysis of the linkage between a certain color and the corresponding mean inbreeding coefficient (chart 5), which shows non-random mating. This shows a significantly higher F (p = 0.014) for red and silver fawn.

 

The comparison of the two ways of calculating the loss of heterozygosity shows a much higher increase of ΔH2 (expected loss of heterozygosity under observed N) up to generation six. From generation six to eight the two values rise equally (chart 4).

 

Dominance of colors: Pivot-table analysis shows more or less arbitrary distribution of the offspring colors based on the color of their parents. None of the colors shows dominance

over another, although red and silver fawn appear to be very common.

 

__________________________________________________________________

 

DISCUSSION

 

Data: The incompleteness of the data which only reaches back to the year of 2003 was a problem for the analysis because dogs that originated from earlier breedings had to be defined as founders. This fact took some influence on the values of the inbreeding- and kinship coefficients by lowering them to some extent. The results can be viewed as explanatory, because the very early generations do not contribute much to the inbreeding- and kinship coefficient of recent animals. The relative change of F does not change with more information about earlier generations.

 

Generation number: The lack of birthdates raised the problem of defining generation numbers, which could be solved to some extent by comparing GnP and GnE. With available birthdates, the average generation interval (L) would be:

 

equ6.JPG(6)

 

Where Lmm is the generation interval between sire and son, i.e. the interval between the birthdates of sire and son, Lmf between sire and daughter, Lfm between mother and son and Lff between mother and daughter, Nomura et al. (2001).

 

By using L, Ne could have been calculated in a more precise way, Nomura et al. (2001).

 

The number of individuals in generation zero from GnE is far too high compared to the actual founder generation (Figure 1) when considering time as an influencing factor. This value shows all individuals with no parents in the record, and thus f = 0 was estimated. These values allowed better evaluation of inbreeding coefficients for the biological

generations, because of the overlapping generations.

 

Random- / non-random mating: Difference between the simulated random mating and applied non-random mating / selective breeding (chart 3) shows that the breeder could have been able to avoid such an increase of the inbreeding coefficient by imitating random mating.

 

The preference of pairing red and silver fawn colored animals could have increased their degree of inbreeding, and thus the probability that the offspring has two identical alleles at one locus. Another possible explanation could be that alleles for red and silver colors show up due to an increased F (increased homozygosity), without breeding preference or due to popular sires. This can eventually lead to an inbreeding depression, which means that fitness reducing alleles show up more frequently in the phenotype, because they can not be hidden in a diploid dominant / recessive system.

 

The question that also has to be addressed is why the curve of the inbreeding coefficients is being flat up to the generation five, although there are bottlenecks in generation three as well as in generation four. Our analysis (data / results) suggests that the new introduced dogs (F = 0) reduce the mean inbreeding coefficient.

 

This can also be seen in chart 4, showing a comparison of different ways of calculating the loss of heterozygosity. ΔH2 increases much earlier (from GnE = 0), whereas ΔH1 begins increasing only from generation six. Because the population size is fairly small, the expected loss of heterozygosity under observed F (ΔH1) should be similar to the expected loss of heterozygosity under observed N (ΔH2).

 

Because this is not the case, we can assume that the effect of newly introduced dogs leads to a lower loss of heterozygosity in generation eight (loss of heterozygosity = 20 %) than expected (loss of heterozygosity = 50 %).

 

After generation six, no new dogs have been introduced into breeding except one which however had no effect because its offspring was so far not included into further breeding. For this reason and due to popular sires, ΔH1 starts to increase.

 

_______________________________________________________________

 

CONCLUSIONS

 

Introducing new dogs into the breeding program is a good way to decrease the inbreeding coefficients in small populations. If random mating would be simulated the coefficients could be even smaller. The fact that the breeding dogs have to fulfill certain requirements concerning their look and / or character makes random mating in a breeding program impossible, when selecting for certain traits. To keep the breed of the American Indian Dog the way it used to be thousands of years ago, selective breeding is required. Our suggestion is a balance between considering coefficients into the breeding program and selective breeding for traits.

 

Further analysis should be conducted using also the breeding data of Kim LaFlamme, reaching back for almost 50 years, to see their impact on the presented results. These data are not available in an electronic form and, due to the time limitation, it was not possible to include it in the current analysis.

 

Ahead of all breeding theories the most important thing is to produce physically and mentally healthy dogs that can be easily integrated into modern society. As shown in Merola M. (2002) inbreeding coefficients can be high over a certain time period without disadvantageous fitness effect if genetic divergence will be reintroduced i.e. through the introduction of unrelated animals.

 

Knowing the breed of American Indian Dog, it becomes clear that it fulfills these crucial requirements and thus is an outstanding and special companion to every owner who knows how to handle this primal breed with strong instincts and diverse abilities.

 

_______________________________________________________________

_______________________________________________________________

AKNOWLEDGEMENTS

 

- Kim LaFlamme, Founder/Trustee of the American Indian Dog Breed and President of the IIDOBA (International Indian Dog Owners and Breeders Association)

- Mike Bloodgood, Founder and President of the IPDBA (International Progressive Dog Breeders Association)

- Dr. Erik Postma, University of Zurich

- Prof. Dr. Lukas F. Keller, University of Zurich

- Dr. Paul Jesenak, ETH Zurich

- Monica Santana, dipl. designer HGKZ

 

 

REFERENCES

- Brinks, J. S., R. T. Clark and F. J. Rice 1962, Estimation of genetic

trends in beef cattle. Journal of Animal Science, 20: 903

- Bouzat J. et al. 1998 & Westemeier et al. 1998, Mendelian

Genetics in Populations II: Migration, Genetic Drift and Non-Random Mating. Evolutionary Analysis, Pearson International Eddition, 2007, 223 – 280

- Calboli et al. (2008), Population Structure and Inbreeding

from Pedigree Analysis of Purebred Dogs. Genetic Society of America. DOI: 10.15.35/Genetics. 107.084954

- Kinghorn B. & Kinghorn S. 2005, Pedigree Viewer Application,

http://www-personal.une.edu.au/~bkinghor/pedigree.

htm, November 2008

- Maki K. et al. 2001, Population Structure, Inbreeding Trend andTtheir Association with Hip- and Elbow Dysplasia

in Dogs. Animal Science 217 - 228

- Malecot G. 1948, Les mathe´ matiques de l’he´ re´ dite´ . Paris: Masson

- Merola, M. (1994). A reassessment of homozygosity and the case for inbreeding depression in the cheetah (Acinonyx

jubatus): implications for conservation. Conservation Biology, 8: 961–971

- Nomura et al. 2001, Inbreeding and Effective Population Size of Japanese Black Cattle. American Society of Animal

Science, 79: 366 - 370

- (2) Mendelian Genetics in Populations II: Migration, Genetic

Drift and Non-Random Mating. Evolutionary Analysis,

Pearson International Eddition, 2007, 223 – 280

- (3) Stalder, Kenneth J. & Saxton, Arnold M. 2004. More estimation of genetic parameters. In: Genetic analysis of complex traits using SAS, Saxton, Arnold M., ed., SAS Institute. Cary, NC

- (4) & (5) Wright S., 1931, Mendelian Genetics in Populations

II: Migration, Genetic Drift and Non-Random Mating.

Evolutionary Analysis, Pearson International Eddition,

2007, 223 – 280

  • Like 1
Link to post
Share on other sites

Join the conversation

You can post now and register later. If you have an account, sign in now to post with your account.

Guest
Reply to this topic...

×   Pasted as rich text.   Paste as plain text instead

  Only 75 emoji are allowed.

×   Your link has been automatically embedded.   Display as a link instead

×   Your previous content has been restored.   Clear editor

×   You cannot paste images directly. Upload or insert images from URL.

Loading...
×
×
  • Create New...