Inbreeding coefficient

edited May 2013 in Breeding
After reading the Inbreeding/Linebreeding thread in the forum today, i found out a website which calculates the inbreeding coefficient and also the ancestor loss coefficient.

Here is the link:
http://www.czerwonytrop.com/inb/index.php

And this is the result for Tajo, with 9 generations (around 200 dogs are not known):

https://dl.dropboxusercontent.com/u/13657625/Tajo.pdf

The inbreeding coefficient is 2.142% (The website says it should be under 4%)
THE AVK (ancestor loss coefficient) is 40.6% (The website says it should be over 80%)


Comments

  • edited May 2013
    There is no "should be" when it comes to COI and AVK. What is important is the qualitative information -- which ancestors was the dog line bred on, why those ancestors, what positive and negative traits do those ancestors have and regularly reproduce in their offspring, etc. Asking "What is the ideal COI?" is like asking "On which weekday should I carry an umbrella?" it depends entirely on the weather, not the calendar!

    That said, its still a useful metric and tool. Me personally, I don't want to linebreed on more than one ancestor at a time (too many variables to account for) so I do a test mating on the JACA website and make sure that I wouldn't be duplicating any ancestors I didn't intend to.

    I happen to like the JACA calculator better... because you can link to it, because it uses our existing database of dogs (no need to re-enter stuff each time), because it also shows parent relatedness, and because I wrote it. ;) Of course it only works for Japanese Akitas and requires some heavy upfront input if your dog isn't in the database. In such a case, the other tool is more convenient, and I have used it in the past.

    http://pedigrees.akita-inu.com/testmating.php
    http://pedigrees.akita-inu.com/testmating.php?sire=54&dam=676&gens=4&coi=4

    I personally like to see low COI and low AVK, which can be difficult to get in combination. Ancestor loss is not a good thing in my book.
  • edited May 2013
    I will try to add my dog and also his ancestors, for sure the code you wrote makes the life easier compared to the website that I shared.

    Can you suggest me some literature that you use? I'm not intending to become a breeder, but however i'm really interested in extending my knowledge :)

    Also is there any book or reference except the breed standard which describes the best characteristics for the JA? So at least i can try to personally judge Tajo's strong and weak points, since I plan to send him to one show in August :)

    Thank you very much @poeticdragon, as always you provide very useful information!

    PS: Thumbs up for PHP, I used to be a PHP developer before i started working as a Linux/UNIX System Administrator :)
  • I've read that the FCI recommends a COI of no more than 10% for 10 generations for rare breeds.
  • Really? Where on FCI regulations is that?
  • I don't know about that 10%. The FCI Breeding Strategy only says this:
    Mating between siblings, mother to son or father to daughter should never be performed.

    When we talk about COIs people need to remember that they individual is not getting sick just because the COI is high. BUT there is higher risk to get sick with recessive or multigenic diseases. For example with Akitas (both JA and AA) you should try to avoid higher COIs because then you have a chance to get puppies who doesn't get the same risks from both parents (they are heterozygote). When you study COIs there is more meaning in closely happened inbreeding in 5-6 generations, that is the area you need to study and try have different dogs in the pedigree. You also need to check the historical inbreeding what can be very tricky with JAs who all are related to same big matadors in the 90s.

    AVK is really important when we think population structure. It tells you basicly how much genetic variety you have lost in this dogs lineage. Every dog should be different in the pedigree and then you get result 1 (or 100% percent) and when it starts to get smaller, the smaller the genetic diversity is. Even though if you have litter where COI is small you can have litter where the ancestor loss is huge when both of the parents are highly inbred. That is situation what Akita breeders should avoid because we need to try to have large genetic variety in the breed so we can avoid diseases like SA and VKH.

    In the close future all breeders will get new tool to make sure there is high as possible genetic variety in the litters. In Finland Genoscoper is bringing to the market genome testing whit program where you can see the breeds genetic situation and also which dog would be best for your dog genetic wise :) There is much more in that and it will come for sale after the WW dog show in Budapest this summer :)

    But these things are highly interesting! I could give you lots to read but unfortunately those are in Finnish :( Here is atleast one good article about the subject what I have found highly interesting. They have other articles in the same subject what are also really interesting and good to read but this is what I find with fast search :)
    The shallow end of the gene pool
    http://kotisivukone.fi/files/heti.tarjoaa.fi/tiedostot/artikkelit/theshallowend_of_thegenepool.pdf
  • Claire's work on our database has been a huge help for JACA. One of the things, that I would love to see is a "breeding score". Something that takes that information and assigns a score on a 1-20 basis. That way a person doesn't see a line breed and automatically flip out, conversely we don't want that same person thinking that an outcrossing means that their puppy will be fine. After all adding crap via outcrossing to a breeding can cause issues as well. Crap + hybrid vigor = Vigorous Crap. :)
  • @MirkaM - I had picked that up recently from this website:

    http://www.czerwonytrop.com/inb/index.php?full=ok =en

    But I remember reading it elsewhere. I'll try to find a cite closer to an official FCI source.
Sign In or Register to comment.