Genetic COI Calculations - Page 1

This question is for anyone that is familiar with Embark testing and the genetic COI portion of these genetic panels. These were test kits that were not done with the Breeder option, so these dogs aren't on the matchmaker tool to see the expected litter COI option. Sire was 33% genetic COI, dam was 31% genetic COI. Is there any type of calculator that I could use other than the matchmaker tool to see the expected litter COI of that outcome? If not, how would I calculate that myself?

Also, say for example you took a puppy from that pairing above and then bred that dog with one that has a 24% genetic COI per Embark, what would that expected COI be?

Thanks.

Correction: pup in question comes from a sire with 27% genetic COI and dam is 33% COI.

It'll depend on if, and how much, overlap there is between the two pedigrees of the two dogs mated. You could use the mating outcome feature on PDB. It won't give you Embarq COI, but it will provide Hardiman's and Wright's.

If the two dogs aren't on PDB, you'll have to add them and maybe some of their ancestors to the database. As you add ancestors, eventually you'll run into dogs already on PDB.

 

Per site rules, you'll need:

"Dog without Progeny
1. Pups under 6 months old require a Date of Birth and a sire and dam who each require a Breed Book & Registration Number
2. Dogs over 6 months old require a Date of Birth, Breed Book & Registration number and a sire and dam who each require a Breed Book & Registration Number. All additional dogs that you enter should include their Breed Book and Registration number if possible.

Dogs with no kennel name require a date of birth due to no kennel name in addition to a Breed Book & registration number."

https://www.pedigreedatabase.com/community.read?post=948404-pdb-requirements

 

Taking a puppy from this mating and mating him or her to another dog, like the first mating, the COI will depend on if, and how much, overlap there is between the two pedigrees.

 

You must first understand terms like prepotency in dogs and homozegous versus heterozygous genes. The PDB will give you Wrights' and Hardminans' equation.  You can look it up, what it means, on this website

Click to view
The PDB will not give you results based on bloodwork, like Embark does, but it gives you Pedigree based results, not blood work based results.  Let's go into this a bit further.

The Genetic COI as measured by Embark and the pedigree COI are different numbers, and you must first understand how both numbers are derived. Here's some information that can help you:

Inbreeding:  2-2 (common father), Relationship coefficient is 50%, and COI is 12.5%  Notice how 12.5% is 25% of 50%?  So the number of common genes is 50%, but only 1/4th of them are homozygeous.
Line Breeding 3-3 (common grandfather) Relationship coffficient is 25% and COI is 6.25%.  Notice that 6.25% is 25% of 25%?  So in a first cousin mating, the common grandfather is 25% of all genes, but only 6.25% of them are homozygeous
Line breeding 2-3 or 3-2. (The male dog's father is the female dog's grandfather) Uncle to niece or Aunt to Nephew mating. The Coefficient of inbreeding is 37.5% but the COI=25% of 37.5% =9.63%.  

Line breeding 4-4, common great-grandfather.  Second cousin mating. Coefficient of inbreeding is 12.5% (or 1/8th of all genes are common. The COI=25% of 12.5%=0.03125, or 3.125%. 
Line breeding 5-5 common great-great grandfather. Third cousin mating: Coefficient of relationship =6.25%  The COI is 25% of 6.25%=1.56%

Notice how the COI is always 1/4th of the COR, the coefficient of relationship. So the total common genes are the COR, but the homogeneous genes are the COI.  

So for example, let's say the COR is 25%, that means one puppy is 13% of the same genes, another puppy is 17% and another puppy is the full 25%.  In other words, it's a range of what the puppy got, since brothers and sisters are not identical clones and each puppy didn't get the exact same genes from the parents.

The Pedigree COI is a mathematical prediction of risk. It's not what is actually took place, it's only what the range that is possible.   In other words, think of the blood based result as the real thing, and the pedigree result as a mathematical calculation of a range of possibilites.  For example, my girl Lucy von Gonta Haus, is a 3-3 breeding on Jucan von Peroh. According to Pedigree database, it's a 6.25% COI, but when I did Embark results, it showed 21%, much closer to the 25% I expected from coeffiicent of relationship.  But if she was mated to another Jucan von Peroh descendant, the mathematical prediction, is more relevant. You can't get blood work results in the PDB, only Wright's or Hardiman's results.  The blood work based results are only done on living animals and there is a range between Lucy and her brothers and sisters, despite having the same two parents. So that is what is the difference between Embark blood work results, or cheeck swab results versus PDB results.
Think of it like if you were to then plan future matings, how many genes are homogeneous in both parents, and that's the number you actually care about, because that's what determines pre-potency. You can have common genes, but unless they are homogeneous, those characteristics are not passed to the offspring.   I suggest you watch the video to better understand the numbes I'm showing.

Here is a good article on COI considerations. It tends to look at the values of dogs further back in a pedigree and may be helpful in sizing things up: https://www.vin.com/apputil/content/defaultadv1.aspx?pId=11165&meta=generic&catId=29801&id=3848663&ind=23&objTypeID=17

Personally, my inclination has been that without the benefit of controlled mutigenerational line breeding, it isn't worth betting the farm on.   Here's and example (Lloyd C. Brackett) if you're unfamiliar: https://breedingbetterdogs.com/article/brackets-formula   Even Brackett apparently had to employ lots of culling to perfect his method.   

I would be interested in your experience of using Embark COI service and how it works out for you.  Please post later on if you go that route, and I'd be interested in others thoughts on the first article I link to above.