Health Assessment in the Mudi Using GCOI

 

A new study was published last week, The effect of inbreeding, body size and morphology on health in dog breeds (link below), which included a bit of data for the Mudi.

Health data was collected from Agria pet insurance, mostly based on dogs insured in Sweden, while “the heterozygosity values were obtained from worldwide sample collection centered in the Scandinavian countries”, using DNA samples sent to MyDogDNA™/Optimal Selection™/(both products of Wisdom Health). They refer to the level of inbreeding used in the study as “genotype-based coefficient of inbreeding”, also known as genetic coefficient of inbreeding, or GCOI.

Number of Mudi used in the study = 141

Heterozygosity minimum = 27.6

Heterozygosity median = 40.4

Heterozygosity maximum = 47.6

Adjusted inbreeding level = 0.096 (9.6%)

Mid range body weight was calculated using AKC breed standards: 10.66 kilos (FCI = 10.5 kilos)

Breeds were sorted into groups based on FCI grouping = No group was listed for the Mudi (although it is listed as belonging to Group 1 in FCI)

Health Insurance Count = none listed

Morbidity = none listed

Brachycephaly yes or no = not listed (the Mudi is not brachycephalic)

The Mudi did not have any health insurance data listed in the data sets, which leaves one to assume they did not have any health data.  The missing health data is not due to the low sample size of Mudis as there were 85 other breeds with a lower count that did have health data (85 breeds have 141 or less dogs and have insurance, morbidity, brachy designation and FCI group information). As a comparison, 222 Pumis and 36 Pulis, were also included in this study and they had insurance, morbidity, brachy data and FCI group listed.

The only mention of the Mudi in the study can be found in this one sentence and Dataset 1:

“The breeds with low inbreeding included recent cross breeds (Tamaskan Dog, Barbet and Australian Labradoodle) and landrace breeds (Danish-Swedish Farmdog, Mudi and Koolie), supporting the notion that high inbreeding is a result of closed stud books or small numbers of founders or both.”

The authors mention landrace breeds once more: “The breeds with the lowest levels of inbreeding were mostly landrace breeds or breeds with recent cross breeding.”

The study states that low inbreeding is less than 10%, the Mudi is 9.6% in their study, if 9.6 is rounded up, it is no longer low, this rounding note will be of some importance later in this post. From the study: “Strikingly few breeds (N = 12) had low inbreeding values (< 0.10)”.

They write that the Mudi is a landrace breed and I agree with that, but so are the Pumi and the Puli.  But the Pumi/Puli inbreeding levels are not below 10%, they are: Puli = 13.6% and the Pumi = 12.1%, according to their DNA sampling. 

They infer that the low inbreeding level in the Mudi is due to the stud book being kept open or from having a high number of founders.  I disagree with the high number of founders existing in the Mudi, it is simply not the case and several of the original founder lines have already been lost, while others were in critical danger of extinction the last time I did a Mudi founder study.

Yes the stud book is still open in Hungary and therefore other FCI countries, but this is not the case already in Canada (CKC stud book has been closed since 2015) and eventually the USA as the stud book closing clock will start to tick January 1, 2022, as that is when the Mudi will become a fully recognized breed with AKC, which means only 3 full generation pedigree Mudis can be registered.  The AKC will leave a window open until 12/2026 for some further registration of less than 3 full generation pedigreed Mudis, via a domestic registry solution, but the details of which domestic registry they will allow are still undetermined, which is quite worrisome.  It is also unknown if they can extend this window after 2026.  This means the Mudi breed will eventually experience an interesting trial between USA/Canada and the rest of the world in terms of inbreeding levels.

Having an open stud book is only valuable if the dogs that are being brought in are truly representative of new or lost lines and the July 23, 2021 Mudi Directions post, The Neglected Hungarian Treasure,  already explained how that has not been the case for quite a long time now.  

So how has the Mudi kept inbreeding levels low?  Are they actually low? Good questions I will research answers to when I have a chance.  I used to publish periodic reports about the levels of inbreeding, but no one seemed to care, so I stopped wasting my time.

I do not disagree that inbreeding levels are connected to health, reproduction and longevity in animals, as many studies have shown it.  As for the other breed data used in the study, I have no quarrel as I am not familiar with their status.

However, inferring the Mudi is healthy because it has a lower level of inbreeding, is not a correct assumption.  The lack of supporting health insurance data is also not an indicator of the health status of the Mudi.  The Mudi data used in this study is not relevant to the purpose of the study.

Regardless of my opinion, I urge you to read the entire study and make your own conclusions.

 

INBREEDING VALUES ARE NOT CREATED EQUAL

I have been recording the Embark GCOI Mudi data for quite a while.  I also tracked the Wisdom/MyDog/Optimal Select GCOI, but there was a problem with the information they provided.  Wisdom does not give a usable GCOI, they give a percentage of heterozygosity (photo example below). Therefore, it is not comparable to my pedigree COI, nor the GCOI that Embark gives which is based on homozygosity.  I can compare the GCOI of Embark to my pedigree database, and my pedigree COI is surprisingly close to Embarks GCOI in most cases I have compared thus far (50+).  

Why do I do this?  It’s more than just curiosity or data collecting, I want to check the reliability of the pedigree COI I give. In many cases Embarks GCOI and my pedigree COI are often within 1-4%. The difference can be due to rounding (I don’t round the COI’s I calculate) and the number of empty spaces on the pedigree.  In a few cases it can be from less common things such as parents being incorrectly reported on pedigrees, accidental breeding with multiple sires, and other breeder/pedigree record mismanagement.

It should be noted that Embark uses the rounding principle with their GCOI’s, for example, if the GCOI is 5.7, they round it up to 6%.  I assume they also round down.

Embark has DNA tested more than 100 Mudis, possibly more than 150.  And these are not just Mudis in the USA or Scandinavia, but Mudis from many parts of the world.  I have just now asked Embark if they can tell me how many Mudis they have tested, if I get a response, I will update this post.

As indicated on a recent Embark results report (see photo at the top of this post), the range of GCOI in the Mudis tested with Embark is between 1% and 30%. What the Mudi breed average is, I cannot tell from the graph. It appears that the bulk of Mudis tested are in the 6-10% range, if the bars indicate that by their height. 

Of course, much depends on the gene set each DNA lab is using as a baseline indicator of homo-heterozygosity for a breed and most likely Wisdom and Embark are not using the same gene set for comparison. Several Mudis have been tested by both labs, with the test results sometimes showing differences, which further indicates testing parameters are not the same.

 

LET’S HERD SOME DUCKS TO ROWS

Even with all the data both labs offer, it is still unknown how representative it is of the Mudi breed in actuality, as a whole. That is, does their GCOI average speak for the whole breed?  Or due to the number and relationship of Mudis tested, is the GCOI average actually much higher or lower?

I had an idea today to see what kinds of statistics I can make using the data I constantly collect.  I decided to compare the GCOI of Embark, to my pedigree COI, to health issues affecting the tested dogs. I also used the rounding function on my pedigree COI to be more in line with the Embark GCOI.

Number of Mudis in my database with Embark GCOI = 50

Embark GCOI range and average of these 50 Mudis: 2-29%; Average GCOI: 11%

USA living Mudis (40) GCOI = 11%; Canada + Europe living Mudis (10) GCOI = 11%

Pedigree COI range and average of these 50 Mudis: 0-21%; Average COI: 8%

Year of Birth/Age range of these 50 Mudis: 2008 to 2021; Aged 13 to less than 1 year

28 Mudis are aged 2 or younger, this means health data is likely not available for these Mudis yet

Country these 50 tested Mudis live in: USA = 40; Canada, UK, Austria, Hungary, Poland = 10

How many of the 50 Mudis produced a litter: 12, between 1 and 4 times

 

LITTERMATE SPREAD IS NOT FOR BREAD

What was also interesting in this GCOI data, is the effect of littermate spread.  

Among this set of 50 GCOI tested Mudis, there are 8 litters containing two or more tested littermates.  The range of the littermates GCOI’s I call littermate spread, the gap is the difference between highest and lowest GCOI of the tested littermates.  For example, the largest set of littermates in this group of 50 is 6 puppies.  Their GCOI’s were 6, 8, 10, 11, 12 (two puppies had 6% each). The spread is 6-12%, with a gap of 6%, which is equal to a 50% difference between the 6% and 12% littermates. That is a significant difference.  However, it is not unexpected for a gap this large to occur in lower GCOI litters.  As the GCOI level increases, it is expected the gap between littermates will decrease.  That is, when the GCOI for a litter reaches the 12.5% and higher range, the gap between littermates GCOI’s should narrow. Why?  Because as the GCOI level rises, the genes shared between puppies in a litter become more and more similarly shared, that is, they have less variation between them, which allows a more similar GCOI.  Less variation in genes creates both good and bad points for a breed, but that is a topic for another post.

In case you are still not sure what littermate spread is and why it is important to understand, think of it this way: the lower the GCOI for a litter, the greater the chance the GCOI for each puppy in the litter to have a very different GCOI from the other puppies in the litter – there can be a big range of GCOI’s among the littermates, this is not unexpected.

As the GCOI level increases, the diversity decreases among the littermates making the differences between puppies smaller – thereby equalizing the GCOI more closely among littermates.

GCOI level is low = higher gap between GCOI’s in littermates

GCOI level is high = lower gap between GCOI’s in littermates

The gap for the 8 groups of littermates found in this group of 50 Mudis, ranges from 0% to 12%, with 5 litters having a 1-4% difference in their GCOI.  The other 3 litters had a gap of 6, 7 and 12%.

As time goes on, it will be interesting to see more of these littermate spreads and if they can be correlated to health issues or longevity between littermates.

 

HEALTH ISSUES AND GCOI

Now for the list you’ve probably been waiting for most.  The following health issues affect these 50 Embark tested Mudis / how many are affected / and the GCOI. Order is alphabetical.

Allergy / 2 / 13%, 15%

Anal Gland Issue / 1 / 7%

Bite: 1 over, 1 under / 2 / 9%, 14%

Distichiasis / 1 / 7%

Elbow Dysplasia / 1 / 7%

Epilepsy / 1 / 9%

Hip Dysplasia / 1 / 14%

Patella Luxation / 2 / 9%, 23%

Teeth Issue / 1 / 9%

Testicles, Missing / 1 / 9%

Thyroid / 1 / 13%

Trichiasis / 1 / 23%

12 health issues in 15 Mudis, GCOI range: 7 – 23%

Percentage of 50 Mudis affected: 30% (15 affected from 50)

Number of affecteds with 10% GCOI or above: 7

Number of affecteds with 9% GCOI or lower: 8

Of course, this is a very small study using only 50 Mudis.  However, a 30% rate of health issues is not small.  It does not appear to be biased towards high or low levels of inbreeding as 7 vs 8 affecteds is not a significant difference.  Perhaps with more Mudis a difference might be seen. 

I do hope to expand this study, however I will need cooperation from the Mudi owners that have done Embark DNA testing.  Please consider sharing your Embarked Mudis results with me.  I will not give subject names to the data, it is anonymous as you can see, and it will always be that way.

If you would like to see the comparison chart between the Embark GCOI and my pedigree COI for these 50 Mudis, I can add it to this post, send me a message that you would like to see it included.

I would like to thank the 49 Mudi owners and breeders that shared their Embark data with me as they made this small starter study possible.  The 50^th Mudi is mine.

If you have any questions, or other statistics or data you would like to see, just ask!

REFERENCES:

Embark report notes, in reference to the chart at the top of this post:

Genetic Diversity and Inbreeding

Inbreeding is a measure of how closely related your dog's parents were. Dogs that are less inbred tend to live longer, healthier lives.

Coefficient of Inbreeding (COI)

Genetic COI measures the proportion of your dog's genes that are identical on the mother's and father's side. The higher the number, the more inbred your dog is.

Wisdom DNA Inbreeding Level Report, photo below:

ARTICLE LINK:

The effect of inbreeding, body size and morphology on health in dog breeds

https://cgejournal.biomedcentral.com/articles/10.1186/s40575-021-00111-4

C.O.I. WHY

We are often contacted with a request to verify the COI (Coefficient of Inbreeding) of different litters as well as give the COI for a planned litter.  Inbreeding is not a guessing game, it is a number that can be quantified if you have the necessary database and formula to calculate it.  Without the proper database and correct calculation program, your COI will be deficient and will lead to inaccurate results, which can have serious consequences for your litter and the breed!

COI is a quick, easy and a free request.  However time is not something we have more and more of these days, so rather than give the COI for the same litter over and over, we have added a new page to this blog that will list the COI for recent and planned litters that we have been asked about.  

If the litter you are interested in is not on this page, you can request we add it by contacting us at:

gonduzomudi@gmail.com

We calculate the COI using every generation we have in the database! Giving the COI for 5 gens is misleading and not relevant to the accurate inbreeding history of a dog or litter.  10 gens is the minimum number of gens recommended by scientists. Why 5 gens has become the 'norm' in certain countries we do not know, but it is not correct to calculate the inbreeding of a litter on only 5 gens.

We also use the known relatives of "B/R" dogs in our calculation when it is known who these parents are that are not listed on the printed pedigree.  

If you have any questions or comments, please be sure to contact us!

Inbreeding 101

COI or inbreeding level, is an integral part of planning a healthy litter and usually the first calculation done when selecting mates.

Inbreeding is both good and bad, it is necessary if you want to make a unique breed of anything, dogs, canaries, peas, whatever.  Some breeders refer to it as inbreeding only when they breed together close relatives such as mother-son, father-daughter or brother-sister, others refer to line breeding which is making sure that a particular dog is in the pedigree of the planned litter as many times as is possible, thereby breeding on that "line" to bring in its traits to your pups and lastly there is out-crossing which means that the sire and dam of the litter are not closely related although they are of the same breed. Inbreeding, line breeding, out-crossing are all inbreeding if the COI is above 0.0%. I personally do not use any of these terms as they are misleading. 

In almost every planned litter, there will be some level of inbreeding.  It is very rare that two dogs with at least a few generations of parents on their pedigree won't have at least one common relative in 10 generations, there are just not that many Mudis hanging out on the puszta these days that are not related to another Mudi that existed somewhere on the planet.  Even Mudis with no parents listed on their pedigree have a once pedigreed Mudi closely related to them in almost every case, and if I know about it, I use it in the calculation of their COI level. 

Your dog and every other dog in existence is either not inbred because their COI = 0.0%, or it has a score above 0.0% which means it is inbred to a lesser or greater degree (COI = Coefficient of Inbreeding, a very long and difficult mathematical equation to figure out how closely related dogs on a pedigree are, don't try this at home without proper math genius supervision). 

In order for a breed to become a breed, some inbreeding has to be done.  Low levels of inbreeding provide the typical fixed basic features of a breed, such as color, fur, body, ear and tail type, and all the other things that make us love a breed and be able to identify a dog as belonging to that breed. High levels of inbreeding over a prolonged period can create more uniform breed specimens. To the not so involved breed person, 10 black Mudis may look all the same, but to those of us that have been forced to gallop around the judge's ring a time or too many, we can usually identify one Mudi from another without too much trouble.  We have only a few Mudi kennels that have "fixed" their "look" on a consistent level to the point that identifying any dog in their lineup from another is problematic for even an experienced eye.  This can be done by either using a persistent inbreeding schedule over many generations or by repeating litters many times over many generations.  Inbreeding is not a quick fix and you can just as easily fix traits you don't want.

The level of inbreeding that is acceptable in a breed depends on a few things, among the most relevant are the breeds average COI level and serious genetic health issue challenges. In some breeds the breed wide COI average is so high, it is not possible to get what is GRAS (generally regarded as safe).  In the Mudi however, we still have a low breed average COI so the GRAS level is under 10%, with 6.25% being the baseline to aim for. 12.5% and above is generally regarded as being high, or higher than GRAS, which actually applies to all breeds, not just the Mudi.  Keeping the COI low should give enough breed similarity to the planned pups, while still allowing for a diverse pool of genes to draw from for future litters, which will benefit health, temperament and the purpose of a breed through the years ahead

So what does all this mean to you?  If your dog has a COI of 0.0%, it is not inbred, this usually happens only when you breed two parents that are not from the same breed together.  If your dog has a COI of 0.1%, it is inbred.  Keeping the level under 12.5% is in the best interest of a breed's future health and welfare. But in order to avoid doubling up the risk of a serious health concern in a planned litter, it may be necessary to go a little bit over the GRAS limits occasionally.  It is probably not wise to go over 16% (a 25% increase from 12.5%), no matter what you are wanting to avoid, the chance of doubling up something additionally not desirable also increases the higher the COI level goes up.

Now if you missed that salient tidbit, the chances of reproducing something good or bad are about equal, no matter the COI level.  That is, if you breed with a lower COI, you might have less of a chance to reproduce something bad, but you also have less chance to reproduce what you want (nice head, tail, fur, etc.).  If you breed with a higher COI you might have a better chance to reproduce what you want, but you also have a better chance to produce what you don't want just as easily (tightly curled tail, missing teeth, genetic disease, etc.).  Now how to rectify this trade off into a win-win situation, that is, you keep the COI level low and still get the good features you want in the planned pups?  

If you know the two parents you plan to use in detail and have studied their visible features well (and have passing health tests for both), if you like what you see, chances are you will like what you see in their pups.  If other dogs that are very closely related to the parents you chose also have these same features, that is, their parents, grandparents, siblings, other offspring they may have had previously, also have curly fur, sickle tail, pricked ears, etc., then the chances get better that your pups will too.  It's an educated gamble, but what you see in the parents and those relatives close to the parents are the most likely choice of features that will be available for the pups in your litter too, regardless of inbreeding levels.  Inbreeding to high levels does not guarantee you will get what you want in your litter anymore than low inbreeding does.

Getting what you want in a litter goes far beyond the inbreeding level!  The inbreeding level is a gauge to help determine the possible gene pool diversity available in a breed, it is not very helpful in getting the features you want in your litter.  As the breed average level goes higher, the dogs in the breed start to show symptoms of inbreeding depression, it is akin to a gauge of health in a breed.  Those breeds that have a high COI average usually have common health and reproduction problems which are difficult to avoid.

So to further put this into action, if you have a bitch you want to breed and you have several males in mind that give a COI below 12.5% preferably, pick the males that will also give you the best chance to avoid major health issues, then from those final choices, pick the males that will improve the faults of your female and have family members that also have those same good qualities.  Not a simple task in such a small breed with males spread out over the globe I know!  I said it was possible to breed better litters, I did not say it would be easy!  But then owning a Mudi to begin with is a challenge isn't it, so don't come to the breed expecting it to get easier!

If you want to learn more about COI's, their calculation, what the percentages mean and what can happen when they get too high in a breed, check out the articles tab on this blog. (coming soon)

If you have any questions or concerns, please send me a private email: gonduzomudi@gmail.com

Thanks for reading!