Breeding Coefficient (sp)

Post Reply
User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Breeding Coefficient (sp)

Post by ohiogsp » Thu Nov 08, 2007 12:11 pm

I am wondering about breeding coefficients (sp). I am breeding 2 of my dogs in the spring. The female is 3 and 4 gen double bred rusty and the male is 4 and 5 gen double bred rusty. Would the coefficient be extremely low?
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

User avatar
Wagonmaster
GDF Junkie
Posts: 3372
Joined: Thu Jun 16, 2005 10:22 am
Location: Minneapolis, MN

Post by Wagonmaster » Thu Nov 08, 2007 12:57 pm

What do you mean by 3 and 4 gen bouble bred rusty. does that mean rusty shows once in the 3rd gen and once in the 4th gen?

there are two ways to look at COI. one, the most common, is a calculation of the general "inbrededness" of the dog. it would look at how many common ancestors show up repeatedly, not just at how often rusty of any specific dog shows up. sometimes, if you have a ped that goes out to say 10 generations, you will get a high COI even if there is not alot of repetition in the first few generations, because there is alot of commonality in the older generations. in GSP's, my rule of thumb is that anything over a COI of 10-12 is getting up there. in some show breeds, the COI's will run to 60, which is way too inbred in my view. but again, to do a COI, you need to put in the whole pedigree, not just a couple of generations or one or two dogs.

the other way is to look at what is the effect of a specific dog. a parent would be 50% of a pup, a grandparent would be 25%, etc. so if you have rusty once in the 3rd gen (great grandparent) that would be 12.5% and once in the 4th gen would be 6.25%, for a total of 18.75% Somewhat less effect than if rusty were one of the grandsires.

User avatar
Theresa
Rank: Champion
Posts: 378
Joined: Fri Mar 30, 2007 12:42 pm
Location: Minneapolis, MN

Post by Theresa » Thu Nov 08, 2007 1:37 pm

You may have to sign up/register but this site says they have an inbreeding coefficient calculator

http://www.tenset.co.uk/fspeed/
Sometimes she wished she were sleeping with the right man instead of with her dog, but she never felt she was sleeping with the wrong dog.

User avatar
ezzy333
GDF Junkie
Posts: 16625
Joined: Fri Aug 20, 2004 3:14 pm
Location: Dixon IL

Post by ezzy333 » Thu Nov 08, 2007 2:11 pm

the other way is to look at what is the effect of a specific dog. a parent would be 50% of a pup, a grandparent would be 25%, etc. so if you have rusty once in the 3rd gen (great grandparent) that would be 12.5% and once in the 4th gen would be 6.25%, for a total of 18.75% Somewhat less effect than if rusty were one of the grandsires
This to me means a whole lot more than any other method of figuring the effect of any and every dog in the pedigree. Not sure how many would agree with me though.

Ezzy
http://www.perfectpedigrees.com/4genview.php?id=144
http://www.perfectpedigrees.com/4genview.php?id=207

It's not how many breaths you have taken but how many times it has been taken away!

Has anyone noticed common sense isn't very common anymore.

User avatar
Buckeye_V
Rank: 5X Champion
Posts: 1373
Joined: Tue Dec 13, 2005 7:48 am
Location: Norwich, OH

Post by Buckeye_V » Thu Nov 08, 2007 2:28 pm

What you need to do is speak with a reproductive specialist or geneticist. They can give you the best information.

What little I know is that I have some dogs that are "line bred." I wouldn't have it any other way. They are proven in the field and competition. They are outstanding companions and thats what I need to know.

Justin
We have done something with nothing for so long we are now qualified to do everything with anything....

http://www.perfectpedigrees.com/4genview.php?id=275

http://www.perfectpedigrees.com/4genview.php?id=520

User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Post by ohiogsp » Thu Nov 08, 2007 3:13 pm

Wagonmaster wrote:What do you mean by 3 and 4 gen bouble bred rusty. does that mean rusty shows once in the 3rd gen and once in the 4th gen?
Yes, once in the 3 on the bitch, once in the forth on the bitch. Then once on the forth of the stud and once on the fifth on the stud.

So, the pups would be like 1 time in the 4th, 2 times in the 5th, and once in the 6th.

So, like johns way would be like 14% right?
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Post by ohiogsp » Thu Nov 08, 2007 3:22 pm

Theresa wrote:You may have to sign up/register but this site says they have an inbreeding coefficient calculator

http://www.tenset.co.uk/fspeed/

According to this site it would be way less the 14% for sure. On this site if the dog is a grandparent on both sides it is only 12.5?
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

User avatar
Wagonmaster
GDF Junkie
Posts: 3372
Joined: Thu Jun 16, 2005 10:22 am
Location: Minneapolis, MN

Post by Wagonmaster » Thu Nov 08, 2007 4:12 pm

Wrights is going at it a little different way than what I gave you. But differences of a few points are not a big deal as far as I am concerned. Nobody is going to do a breeding because it results in a 14 rather than a 12. 14 and 12 are pretty low numbers if you are thinking that you have or want to have linebred pups. Just my own personal rule of thumb, in GSP's I would say a 12 is about the bottom number you want to see, if you are thinking in terms of a linebreeding. Most true linebreedings will be in the range of 16 up to 32 - in modern field bred GSP's.

Frankly, the quality of the dogs you breed is a great deal more important than a couple or even several points in a COI calculation.

User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Post by ohiogsp » Thu Nov 08, 2007 4:34 pm

Wagonmaster wrote:Frankly, the quality of the dogs you breed is a great deal more important than a couple or even several points in a COI calculation.
I was just curous. I think this breeding will produce some very nice dogs and it won't be because of Rusty. Well he might of had something to do with it in the long run.


I was just looking at the bitches 5th gen and rusty is in there 2 more times. This would bring it to 17.187


What is out there close to rusty today. Is there direct offspring left?
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

lvrgsp
GDF Junkie
Posts: 2511
Joined: Sat Jun 10, 2006 9:12 am
Location: ILLA NOISE..................

Post by lvrgsp » Thu Nov 08, 2007 6:54 pm

Direct offspring, I don't know of any living, 1 son who was collected that I know of is Shoulderstrap rusty a NSTRA CH. I know nothing about him. There are some grandsons out there, Kojac's K-man, HMK's Self possesed, G-Gson Diamond S Angus. I know there are more those are just off the top of my head. I would have to believe there are a few more Rexx sons out there as well. Charlie Rose may have something tucked away for a rainy day as well, or he might know of some more.

FWIW,
Chip 8)
Last edited by lvrgsp on Sat Nov 10, 2007 9:02 am, edited 1 time in total.

User avatar
Ayres
GDF Junkie
Posts: 2771
Joined: Sat May 08, 2004 9:01 pm
Location: Flat Rock, IL

Post by Ayres » Thu Nov 08, 2007 9:12 pm

You might want to try this out:

http://mbfs.com/

The Breeder's Standard 2008 is out and you can download a 14-day trial version (Retail $89.99 for full version). It takes awhile to put in the dog information, of course, but when you're done you have a nice database of information to access. You can use TBS 2008 to do a very nice COI calculation, as well as create .pdf pedigrees for prospective puppy buyers.

I think TBS 2008 uses a slightly different way of calculating the COI though. The sire and dam each contribute 25%, gr-sire/dam each 12.5%, etc. Not as high of numbers.
- Steven

Justus Kennels.com

Justus James Ayres SH CGC - Justus - Rest in Peace, buddy.
Wind River's JK Clara Belle - Belle
Wind River's JK Black Tie Affair - Tux

Over the Hill

Post by Over the Hill » Fri Nov 09, 2007 9:44 am

Ohio GSP:

We have a small shorthair database, and my wife calculated your bitch’s inbreeding co-efficient for you. It’s 11.6%. using 8+ generations.

You actually have more Rusty in there than you think. Twice in the 5th, in Stormcrow In Dixie and Rusty Spur, who were Rusty sons, and once in the 6th, in Gotta Go Koojo, who was a Rusty grand daughter.

But LB’s Ohi Cricket and LB’s Ohi Kate were littermates, and up much closer than Rusty. They put in more to the inbreeding co-efficient than he does.

Jim

User avatar
ACooper
GDF Premier Member!
Posts: 3397
Joined: Tue Nov 23, 2004 1:37 pm
Location: Sometimes I'm in Oklahoma

Post by ACooper » Fri Nov 09, 2007 10:05 am

I have a deposit on litter the COI is 24%, and I cant wait.

User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Post by ohiogsp » Fri Nov 09, 2007 10:21 am

Over the Hill wrote:Ohio GSP:

We have a small shorthair database, and my wife calculated your bitch’s inbreeding co-efficient for you. It’s 11.6%. using 8+ generations.

You actually have more Rusty in there than you think. Twice in the 5th, in Stormcrow In Dixie and Rusty Spur, who were Rusty sons, and once in the 6th, in Gotta Go Koojo, who was a Rusty grand daughter.

But LB’s Ohi Cricket and LB’s Ohi Kate were littermates, and up much closer than Rusty. They put in more to the inbreeding co-efficient than he does.

Jim

Thanks, I am going to PM you.
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

Dave Quindt
Rank: 5X Champion
Posts: 876
Joined: Tue Jul 05, 2005 8:22 pm

Post by Dave Quindt » Fri Nov 09, 2007 10:36 am

Let's not forget that COI assumes that the pedigree is 100% correct. Even forgetting possible cross breeding, I have a hard time believing that most pedigrees are perfectly correct over 5+ generations.

I have a database of 3000+ field-bred GSPs and can run COIs until my fingers bleed. For 95% of the folks interested in COI, it is a completely worthless (and misleading) number.

FWIW,
Dave

Over the Hill

Post by Over the Hill » Fri Nov 09, 2007 11:55 am

Ohiogsp:

I could only do your bitch's inbreeding co-efficient because we used to own a dog that was closely related to her, so most of your information was in our database.

Only two of the dog's in your male's pedigree are in our database, so I can't help you out. Sorry.

We have a large dairy goat database, (because we're advisors for a dairy goat 4-H club) a medium sized Border Terrier database because my wife occasionally breeds her BTs, and a small Shorthair database because I'm not a dog breeder, just a hunter who keeps track of a few pedigrees of dogs that are related to the ones I've owned. Sooner or later I'm going to be looking for another dog, and I think the breeding is important.

In most cases, I'd have to agree with what was said above about inbreeding co-efficients being useless for most people. It's much more useful with the goats where milk production and butterfat records are available and accurate for all the animals leading back to the common ancestor(s).

If you aren't personally familiar with all the important characteristics of all the animals in the direct line of descent an inbreeding co-efficient is meaningless, and if you are, you won't need it!

Good luck with your breeding,

Jim

[/quote]

User avatar
ohiogsp
Rank: 5X Champion
Posts: 1238
Joined: Tue Nov 16, 2004 8:13 pm
Location: Toledo Ohio

Post by ohiogsp » Fri Nov 09, 2007 12:41 pm

Cool, thanks for your help.
<table width="300" border="0" cellspacing="0" cellpadding="4"><tr><td width="75"><a href="http://www.perfectpedigrees.com/genview ... =184"><img border="0" src="http://www.perfectpedigrees.com/picture ... /td><td><a href="http://www.perfectpedigrees.com/genview ... 184">DIXIE HIGHWAY'S BOOZE RUNNER JH
<a href="http://www.perfectpedigrees.com"><font size="2">Get your free pedigree!</font></a></td></tr></table>

anne
Rank: Junior Hunter
Posts: 70
Joined: Wed Mar 28, 2007 4:27 pm

Post by anne » Fri Nov 09, 2007 1:35 pm

Also what needs to be considered is over how many generations is the COI being calculated? To be consistent, it's best IMO to run COIs with your own database, there's not point in having a given number and comparing it to someone else's number if the numbers are run from different databases. Also most people don't have very complete databases, and of course Dave also makes good point about accuracy of databases.

User avatar
ezzy333
GDF Junkie
Posts: 16625
Joined: Fri Aug 20, 2004 3:14 pm
Location: Dixon IL

Post by ezzy333 » Fri Nov 09, 2007 1:41 pm

For the sake of educating everyone on the board would someone explain COI and what it tells you?

If it a statistical forumla why are there different ways to figure it?

Thanks,

Ezzy
http://www.perfectpedigrees.com/4genview.php?id=144
http://www.perfectpedigrees.com/4genview.php?id=207

It's not how many breaths you have taken but how many times it has been taken away!

Has anyone noticed common sense isn't very common anymore.

Over the Hill

Post by Over the Hill » Fri Nov 09, 2007 3:46 pm

There are two different ways to calculate an inbreeding co-efficient: Wright's Equation (1921) and Hardiman's Method (2000).

Both show the mathmatical probability of an animal being homozygous for any certain gene pair.

The biggest difference is that Wright's Equation only considers ancestors that occur in both parent's lineage, while Hardiman's Method considers the degree of inbreeding of each parent.

The first problem with both is that they start with the assumption that each ancestor included in the calculation was homozygous for the gene pair being considered. This is almost never true in dogs, so that's the first big hole in both.

The second is that both methods are based on random selection. They ignore the fact that someone is selecting for certain traits and against others.

We use Wright's Equation, because we were using it before Hardiman's Method was introduced. It's likely less accurate but it works well enough for the two things the kids are trying to keep track of in their goats.

I'll try to come up with a couple examples of how it actually works after dinner tonight, and get some input from my wife. She's a biology teacher and knows a lot more about this stuff than I do.

User avatar
kninebirddog
GDF Premier Member!
Posts: 7846
Joined: Sat Dec 25, 2004 12:45 am
Location: Coolidge AZ

Post by kninebirddog » Fri Nov 09, 2007 6:44 pm

www.mbfs.com

You can download a full version which you can use free for 14 days

enter all information you have and can get

You can get a COI and also a Ancestor contribution report
"I pledge allegiance to the flag of the United States of America, and to the republic for which it stands, one nation under God, indivisible, with liberty and justice for all."
"When I hear somebody talk about a horse or cow being stupid, I figure its a sure sign that the animal has outfoxed them." Tom Dorrance
If you feel like you are banging your head against the wall, try using the door.

Over the Hill

Database Software

Post by Over the Hill » Fri Nov 09, 2007 7:12 pm

The free trial is nice, but entering animals in any database is a lot of work. Once you've spent a lot of time on it you're going to buy their software or lose all the work you put into it.

I'm not familiar with the "Man's Best Friend Software" but one thing to watch for is whether you can export/import to and from other databases. Some lock you in to their software while others allow you to convert your data to other programs.

User avatar
kninebirddog
GDF Premier Member!
Posts: 7846
Joined: Sat Dec 25, 2004 12:45 am
Location: Coolidge AZ

Post by kninebirddog » Fri Nov 09, 2007 7:30 pm

pretty easy to enter they do a BXF format but after trying some other programs I tried MBFS and turned around less then 15 minutes i was buying it have been using it now for over 4 years
"I pledge allegiance to the flag of the United States of America, and to the republic for which it stands, one nation under God, indivisible, with liberty and justice for all."
"When I hear somebody talk about a horse or cow being stupid, I figure its a sure sign that the animal has outfoxed them." Tom Dorrance
If you feel like you are banging your head against the wall, try using the door.

User avatar
original mngsp
Rank: 5X Champion
Posts: 1232
Joined: Fri Dec 16, 2005 1:24 pm
Location: Mandan, ND

Post by original mngsp » Fri Nov 09, 2007 9:19 pm

I have never bred any liters but have a pretty good grasp on the idea of COI.

My question here is. Since the higher the COI is, it generally means that the probablity of traits and other things effected by the gentics will be increased, in other words increasng predicibilty. We all have ideas of what traits we would like to purpetuate and what we would like to decrease.

So my question is..... how does a breeder know what good things will carry forward vs. what bad things will show up. It seems to me that try as we might it still isn't a 100% science by any means. Does this leave the breeder to use his or her knowledge, what available science(COI) is available, and even then still are somewhat rolling the dice and hoping to come out the right way.

So in other words it is still basically an art that we are starting to use science to improve the artistic vision.

User avatar
nrcgsp
Rank: Master Hunter
Posts: 243
Joined: Fri Apr 27, 2007 8:47 pm
Location: Missouri

Post by nrcgsp » Sat Nov 10, 2007 1:05 am

I use the mbfs program and really like it! You can do "what if" breedings, print pedigrees, vet records, etc. entering dogs can be time consuming, however, if you find someone who has a verison that yours will upload/import from they can even email you a file. We raise GSP's our national champion that we lost in 2006 was a Rusty grandson, we have a male out of him now (so a great grandson to Rusty). I just sold a female who was out of Magnum also, hard to find them much closer than grand-get unless semen has been drawn on them.
If you use the mbf software, I could try to export files I have (several gsp pedigrees and litter pedigrees). Good luck w/the breeding.
Nancy
Home of: NFC/FC/RegCh/NRC'S Magnum's Touch of Gold(We miss you Magnum);
NSDC/FC/Reg. R-upSlick Kessie Joe; http://www.perfectpedigrees.com/4genview.php?id=598

lvrgsp
GDF Junkie
Posts: 2511
Joined: Sat Jun 10, 2006 9:12 am
Location: ILLA NOISE..................

Post by lvrgsp » Sat Nov 10, 2007 9:13 am

Ohio, a few more Rusty g-sons, really cannot believe I missed them.
Dunfur's Joe Obvious (Simon), Diamond S texas Express (Tex), not sure if he is still alive and breedable though. He is a product of a Rusty Son bred to a Rusty G-daughter, a nice linebreeding on Rusty, and a brother to Stevie, Diamond S Stephanie.

FWIW,
Chip 8)

Over the Hill

Post by Over the Hill » Sat Nov 10, 2007 9:55 am

original mngsp wrote: ..... how does a breeder know what good things will carry forward vs. what bad things will show up.
I got a private message asking almost the same thing, but also wondering why if we think COIs are almost useless, we encourage a group of 4-Hers to use them.

With goats, we don't care how fast or far they run, how easy they are to train, how good a "nose" they have, how they carry their tail, etc, etc. All we care about is how much milk they produce, the amount of butterfat in the milk, and milk production average over several months. Part of each kid's 4-H project involves keeping accurate records of these things, which are entered into our database. The important qualities in a goat can be reduced to statistics, while the things we look for in a bird dog can't.

Another thing that gives value to COIs in goats is that the bucks don't produce milk. You're sort of "shooting in the dark" when you decide to breed to an unproven sire. The statistical probabilities that he's inherited the things you want from his ancestors are really the only thing you can use to decide if you want to breed him. (Imagine trying to breed bird dogs when you have no way of knowing whether the sires available can hunt.)

All a COI tells you is the PROBABILITY that an animal has inherited double genes from a certain ancestor, one coming down through the sire line and the other through the dam's. (The proof is in the quality of his daughters.)

The actual number (percentage) tells you absolutely nothing. Not which trait may have been inherited (good or bad) or even which specific animal it may have come from. Like I told Ohiogsp when I ran a COI on his bitch, the two LB's Ohi bitches contributed much more to her inbreeding co-efficient than Rusty did. If she's an exceptional dog, it's more likely (statistically) that her quality came from these bitches.

It's even more likely that some genes coming down from Rusty combined with some genes from the LB bitches to produce the result. (good or bad) All this really does is let the breeder make an educated guess: breed to 1) a male that has much of the same background from both lines; 2) a tightly linebred Ohi male; or 3) a tightly linebred Rusty male.

A lot depends on exactly what the breeder's idea of a good dog is and what he hopes to produce, which is another problem with trying to use COIs to plan breedings. Ohiogsp's bitch might have a great nose, but be hard to control. If he made all three of the breedings with her, and kept track of ALL the offspring from each, he'd start to get an idea where these traits (good and bad) were coming from. He'd also want to get to know as many dogs as possible from very similar breedings. After keeping track of these and a few generations of selective breeding, he'd begin to be able to use a COI to help predict the kind of dogs a certain breeding should produce.
original mngsp wrote:So in other words it is still basically an art that we are starting to use science to improve the artistic vision.
I'd say you got it right!


Unfortunately in dogs COIs don't seem to be much more than advertising hype, implying "My litter has a low COI so there won't be any genetic defects," or "My litter has a high COI and goes back to Ch. World's Greatest Hunter 27 times, so the pups will be just like him." Neither is true; just at best a statistical probability, and can't be inferred from a COI.

COIs in dogs would probably be most useful in pin pointing the source of identifiable genetic defects, such as hip dysplasia. But to do even this, you'd need a database with hundreds of thousands of dogs, all of which had been x-rayed.

Over the Hill

Post by Over the Hill » Sun Nov 11, 2007 9:52 am

Inbreeding co-efficients are very easy to understand, but they're awfully hard to explain without a few diagrams and visuals.

I have a Power Point presentation that we use for 4-H, and 10-18 year old kids seem to understand it very easily. I'd be happy to share the parts that would apply to dogs if there is a way to use a few diagrams here.

My question is this: Is there a way to get pictures from my computer to this board, and if so, how?

User avatar
kninebirddog
GDF Premier Member!
Posts: 7846
Joined: Sat Dec 25, 2004 12:45 am
Location: Coolidge AZ

Post by kninebirddog » Sun Nov 11, 2007 10:07 am

On your key board there is a PrtSc SysRq key upper left
bring up the screen you want to show full size then press that key

in your paint program open that then paste the picture

besure when saving where you title just below is another window save in the Jpg jpeg format

then down load to a phothosting program like Photobucket

here is where i am showing people how to post a resizable thumbnail from photobucket
Image
Image
click to enlarge
Image
"I pledge allegiance to the flag of the United States of America, and to the republic for which it stands, one nation under God, indivisible, with liberty and justice for all."
"When I hear somebody talk about a horse or cow being stupid, I figure its a sure sign that the animal has outfoxed them." Tom Dorrance
If you feel like you are banging your head against the wall, try using the door.

Over the Hill

Post by Over the Hill » Mon Nov 12, 2007 9:16 am

For those who are interested in COI, I posted a pedigree of

<a href="http://www.perfectpedigrees.com/genview.php?id=878">NFC Moesgaard's Angel's DeeJay</a>

and included the inbreeding co-efficients for each ancestor, calculated using Wright's Equation.

Her pedigree is the most intensely linebred that I've ever seen in Shorthairs: A double half sib breeding that included 3 generations of father to daughter breedings on each side.

If there are no other common ancestors, two generations of father x daughter breedings, using Wright’s Equation, would yield a COI of 37.5%. It’s higher on Angel’s pedigree, because Ib and Bunny did share some common ancestors, as evidenced by the increased COI in Lucky.

According to my wife, two generations of father to daughter breedings (again, assuming there were no other common ancestors), calculated using the Hardiman Method, would yield an inbreeding co-efficient of 78%.

In a post above Wagonmaster mentioned dogs with ICOs of 60% and higher. I assume these were calculated using Hardiman's method, because the type and degree of inbreeding needed to reach 60% using Wright's Equation makes it highly unlikely.

To test this, I did several theoretical breedings. (In other words, I invented some dogs and did pedigrees and co-efficients for them.) Here are the results:

Angel (45.1%) x Ib (7.8%) = Candy (45.3%)
Angel (45.1%) x Coco (43%) = Dolly (58.3%)

Then, I “created” a full sib to Angel, and labeled him “Axel,” and progressed to tighter and tighter hypothetical inbreedings from there.

Angel (45.1%) x Axel* (45.1%) = Baron & Bella (58.3%)
Baron (58.3%) x Dolly (58.3%) = Echo (61.9%)
Bella (58.3%) x Ib (7.8%) = Gretta (45.6%)
Echo (61.9%) x Ib (7.8%) = Hilga (45.6%)
Baron (58.3%) x Bella (58.3%) = Ivan & Irma (65.4%)
Irma (65.4%) x Ivan (65.4%) = Junker & Juno (72.3%)
Junker (72.3%) x Juno (72.3%) = Karl (77.5%)

Then I backed up, and “created” a bitch littermate to Ruffy (already 3 generations of father to daughter breedings on Ib) and labeled her “Rose.”

Rose (42.9%) x Ib (7.8%) = Sally (48.4%)
Sally (48.4%) x Ib (7.8%) = Tess (51.1%)
Tess (51.1%) x Ib (7.8%) = Ulk & Unna (52.5%)
Ulk (52.5%) x Unna (52.5%) = Vex (58.6%)

As you can see, 6 generations of father to daughter breedings, calculated using Wright’s Equation, yielded a COI of 52.5%, compared to 2 generations yielding 78% using Hardiman’s method.

The only way I could bring a COI to (nearly) 78% using Wright’s Equation was through 6 generations of full sib breedings (77.5%).

Now here’s where it gets interesting. Ib accounts for virtually all of the COI (using either method) in the breedings above. He was a roan tick-patched dog, and the co-efficients predict a very high probability that Unna or Karl (the end results of the two forms of inbreeding) will be roan tick patched dogs.

Assume for a moment that Bunny was a solid liver (she wasn’t!) and that the person doing these breedings (now we’re looking at 10 generations—Angel’s pedigree, + the six hypothetical generations.) selected a solid liver in each generation.

In regard to this particular (color) gene, Unna (COI 52.5% --father to daughter breedings) in reality is no more likely to be a roan tick-patch than an out crossed daughter would be, or 50%.

Karl, with a COI of 77.5% (full sib breedings) and two solid liver parents has only a 25% potential of being a roan tick patch, and MAY be completely unrelated to Ib for this gene, if at any point in the preceding generations the gene was lost.

It’s common for an animal to be genetically unrelated to a grandparent for any particular gene. It’s possible, though highly improbable, for it to be completely unrelated. A COI of 12.5% tells you that of the animals with this exact pedigree, 1/8 of them are likely to have inherited a gene from Ib. It absolutely does not tell you what your individual animal actually has inherited.

Another important thing to bear in mind: If a CIO of 12.5% tells you that 1/8 of the dogs with a certain pedigree probably inherited certain genetic material (one in a litter of 8) it also tells you that 7/8s (the other 7 puppies) probably didn’t!

Post Reply