Poodle Forum banner

1 - 16 of 16 Posts

Super Moderator
Joined
3,140 Posts
Discussion Starter #1 (Edited)
Two puppies that share the same father but different mothers are half-siblings. That pup would have a too-high COI of 25%.

This situation is a little different. I'm considering a line bred puppy where Dog X is the father of the sire and the grandfather of the mother.

All things considered, what is the approximate COI of this puppy?


I got this COI info below from here.

100% identical twins; clones
50% parent-offspring
50% full siblings
37.5% 3/4 siblings or sibling cousins
25% grandparent-grandchild
25% half siblings
25% aunt/uncle-nephew/niece
25% double first cousins
12.5% great grandparent-great grandchild
12.5% first cousins
12.5% quadruple second cousins
9.38% triple second cousins
6.25% half-first cousins
6.25% first cousins once removed
6.25% double second cousins
3.13% second cousins
0.78% third cousins
0.20% fourth cousins
 

Registered
Joined
919 Posts

Super Moderator
Joined
3,140 Posts
Discussion Starter #8
I'll tell you what's interesting. I used the link from SMSP but did four generations: the parents, grandparents great grandparents and great-great grandparents. That came up with 12.5% COI for the pup.

However, when I did five generations, the COI is 6.25%. The more generations you do, the lower the reported COI. I'm guessing this is the same software algorithms used by poodledata and others.

They don't have available for three generations: parents, grand parents and great grandparents.

[Btw, there's a shortcut to this so you don't have to fill out the name of every dog in the pedigree. For unrelated dogs in filling out the form, you can name them A, B, C, etc, or even leave those blank and input only the Puppy and the related dogs.]

These COI tools are nice, but for dogs with relatives in the three most recent generations, i.e., parents, grandparents and great grandparents, I wonder how useful any of this kind of software is.

This takes us to Scooterscout99's link to the UCDavis Genetic Diversity Testing for Standard Poodles.

It's marvelous from what I've read about it in the past month, but not something you can learn how to interpret in 15 minutes unless you just have that skill or background. There is a free mini-course here. And here's one of several videos by Dr. Rebekah Zurbrugg who is affliated with BetterBred and UCDavis.

 

Registered
Joined
1,213 Posts
This is why many are turning to the VGL testing.

Phin -
Poodledata:
10-generation COI 0.31% 12-generation COI 1.18%

Monster
Poodledata:
10-generation COI 5.73% 12-generation COI 6.20%

Using the diversity testing...

Phin
BetterBred
Outlier Index (OI): 0.15 Breed Average OI: 0.24
Average Genetic Relatedness (AGR): 0.05 Breed Average AGR: -0.02
Percent of Dogs Unrelated (PCUR):33.81 Breed Average PCUR: 57.08%

Monster
Outlier Index (OI): 0.26 Breed Average OI: 0.24
Average Genetic Relatedness (AGR): -0.09 Breed Average AGR: -0.02
Percent of Dogs Unrelated (PCUR): 77.71 Breed Average PCUR: 57.08%

I could only keep one for use as a stud. Can you guess which one and why he remained here?
 
  • Like
Reactions: scooterscout99

Super Moderator
Joined
3,140 Posts
Discussion Starter #10
Spindledreams, your question at the end was an excellent exercise. Without peeking first at the BetterBred site, I guessed Monster b/c he has a higher than Outlier Index (OI). However, I couldn't coherently why into words; I just knew it was a good thing.

So I refreshed myself by looking at the video again.

A Standard Poodle has this (link):

Breed Average Outlier Index: 0.24
Breed Average Genetic Relatedness: -0.02
Breed Average Internal Relatedness: 0.03

(See this link for other breeds and Toy Poodles tested.)

There was a mid-century genetic bottleneck where too many Spoos were inbred, resulting in less genetic diversity among them. More breeders now are trying to get away from inbreeding and line breeding as a result.

See video in Post 8, starting at 1 Minute-45 Seconds. The higher Outlier Index (OI), the more you're breeding away from a bottleneck; the higher the breed average, and the you'll see fewer breed specific health problems.


Your dog Monster

Outlier Index (OI): 0.26. The Breed Average OI: 0.24
Average Genetic Relatedness (AGR) in the negative numbers at -0.09, which is better than the breed average of -0.02.

Contrasted to Phin, Monster has more genetic diversity and more breeding options being less likely to contribute to the genetic bottleneck.

I still don't know this stuff well enough to be eloquent in describing it. If you can explain it better I'd appreciate that.
 

Registered
Joined
1,213 Posts
The Facebook group has a handy cheat sheet to help learn what you are looking for.

diversity cheat sheet by spindledreams, on Flickr

Yes I did pick Monster for his higher OI giving more breeding options if I pick a mainstream girl like Twilight for his mate. Monster does have a much higher than average IR ie his is 0.31 and the breed average is 0.03 so I want to lower that in his offspring but he is the product of a recent line breeding.

The program showed that a breeding between Monster and Twilight would INCREASE the average OI of the puppies from Twilights 0.15 to 0.21 which is a pretty big jump. That breeding would also drop the IR (how inbred the dog is) from Monsters high of 0.31 to an average of 0.06 giving each puppy a greater diversity of genes.

If I was using Phin and Twilight the OI remains at 0.15 but the IR increases in the puppies. I have gotten pretty similar results in any of the test breedings I did on BetterBred with random bitches. So I felt that despite his high IR Monster would be the better pick of the two. Giving me a higher OI in most litters and if his mates were picked carefully I could lower his higher than average IR.
 

Registered
Joined
1,213 Posts
Also please note that had I used only the COI from Poodledata in my decision I would have picked differently. On paper Monster is much higher then Phin. And notice Cole my younger male who is actually 1/4 Mini shows a lower COI then Monster but higher then Phin...
Monster : 10-generation COI 5.73% 12-generation COI 6.20%
Phin 10-generation COI 0.31% 12-generation COI 1.18%
Cole 10-generation COI 0.86% 12-generation COI 1.40%
 
  • Like
Reactions: Vita

Super Moderator
Joined
3,140 Posts
Discussion Starter #13
Wow, this cutting edge, state of the art genetics; I have a deeper appreciation for it after seeing your real life example:

Also please note that had I used only the COI from Poodledata in my decision I would have picked differently. On paper Monster is much higher then Phin. And notice Cole my younger male who is actually 1/4 Mini shows a lower COI then Monster but higher then Phin...
Monster : 10-generation COI 5.73% 12-generation COI 6.20%
Phin 10-generation COI 0.31% 12-generation COI 1.18%
Cole 10-generation COI 0.86% 12-generation COI 1.40%
Tell me, if Person A has a stud and person B has a female, and both have the diversity test, how do you or is it possible to find out before a mating what the general results of the puppies would be? Or do you have to test each pup individually? I might get a pup or just breed Bella in the future so I'd like to know, and this would be helpful to others. Thanks.
 

Premium Member
Joined
22,964 Posts
Vita I am a little surprised to see that you are still considering breeding Bella. If you do think you are serious about that please get an experienced mentor who will be honest with you about the risks and benefits as well as doing an objective evaluation of Bella's potential for adding to the quality of the breed. I know you care about our breed and that Bella is the light of your life (children aside), but there are lots of great breeders out there to help newbies and think their brains should be picked. It isn't just about the genes but what kind of dog those genes put together in terms of structure, coat, temperament, etc. I adore my dogs personally and I know a lot about dogs from many perspectives, but I don't think I am anyway qualified to breed without some very good mentoring.
 

Registered
Joined
1,213 Posts
IF you are a member of BetterBred you can select any public dog to do test matings with your dog. The test mating function uses at least 500 puppies to produce an analysis of what is possible. The test mating function also lets you look at possible mates based on each of the following criteria:

Potential Breedings by Genetic Value to the Breed (OI)
Potential Breedings by Similarity to the Breed (AGR)
Potential Breedings by Genetic Similarity Between Mates (GR)
Potential Breedings by Color

So you can actually pick what is important to you or look at more than one list to find a male you are interested in.

This is just one tool in our tool set now for selecting pairs for breeding, you still need to make sure both are a good physical and temperament matches just as has been done in the past.

Also while I posted averages in my sample the test breeding will also show you the whole range that could happen. For instance in the Monster x Twilight breeding here is more information from that test breeding:
Average Outlier Index 0.21
Minimum to Maximum OI 0.11 to 0.38

Average Internal Relatedness 0.06
Minimum to Maximum IR -0.20 to 0.32

Here are the numbers for the Twilight x Phin test breeding
Average Outlier Index 0.15
Minimum to Maximum OI 0.06 to 0.26

Average Internal Relatedness 0.09
Minimum to Maximum IR -0.17 to 0.43

Look at the differences!! Any pups I was thinking of keeping would have to be tested to see where they fell in that range and it would become one more tool in my selection criteria.
 
  • Like
Reactions: Vita

Registered
Joined
1,439 Posts
I enjoyed reading this as I will be getting the Vgl test done on Luna soon. No idea what to expect from it as she will be the first one in her line (that I know of) to be done.
 
  • Like
Reactions: Vita
1 - 16 of 16 Posts
Top