Mating a half brother to a half sister.

[southern cockers]

[quote][/quote]mateselect hasnt invented anything new here, its just opened up to all an easy way to get to as accurate a COI and all historical health info as is available at present for a given dog or mating.[quote]

Let me state at this juncture that I have long been a proponent of and derived immense satisfaction from the research of pedigrees using Wright's C.O.I. and the associated ancestral % loss long before the K.C decided to adopt such a programme and as such I welcome their efforts .
However it is absolutely INCORRECT to state that the readings from the K.C Mate Select programme are in any way ACCURATE as like is not being compared with like. In fact they are quite misleading save for singular dogs C.O.I.
If someone can explain to me that there is no discernable difference in the calculations of an index derived over 9 generations as applies to some readings compared to 15 generations that applies to others I will take the next plane to Mars !
I welcome the programme and its associated benefits in assessing the virtues of any potential mating but until such time as all dogs are equally assessed and a common denominator of generations applied in each instance the index/ coefficient is virtually worthless !

[Millburn]

Hey
Just what will you do with all your spare time now southern as the way I see it the kennel club have kinda put you out of a job. I am just even more confused now.

[arwesa]

mateselect hasnt invented anything new here, its just opened up to all an easy way to get to as accurate a COI and all historical health info as is available at present for a given dog or mating.

Let me state at this juncture that I have long been a proponent of and derived immense satisfaction from the research of pedigrees using Wright's C.O.I. and the associated ancestral % loss long before the K.C decided to adopt such a programme and as such I welcome their efforts .
However it is absolutely INCORRECT to state that the readings from the K.C Mate Select programme are in any way ACCURATE as like is not being compared with like. In fact they are quite misleading save for singular dogs C.O.I.
If someone can explain to me that there is no discernable difference in the calculations of an index derived over 9 generations as applies to some readings compared to 15 generations that applies to others I will take the next plane to Mars !
I welcome the programme and its associated benefits in assessing the virtues of any potential mating but until such time as all dogs are equally assessed and a common denominator of generations applied in each instance the index/ coefficient is virtually worthless !

If you go re read the mateselect thread. I took the time to research the way COI's are calculated, the algorithm the KC are using and what happens when it doesnt have the data for past breedings. fact is the more gens you include in a COI the more accurate it is...its a guide, so using a predetermined cut off of 10 gens and then assuming theres no inbreeding prior to that (as you would with your software) is even more misleading...

As I understand it, the COI is a guide to tell you with as much accuracy as possible how much inbreeding there is for a particular dog its not a score out of 100 based on the last x generations. ideally to get a tru COI you need all the breeding info for a dog from the day the breed existed but obviously thats just not possible, so you use as much info as you can and have to assume where you dont have info prior to that there was no inbreeding (i.e. assume the best)...not ideal but thats the way it is. giving you how many gens they are going back and whats the shortest number of gens they have on record just gives your more discerning user a guide to how accurate the score is vs another dogs COI for comparitive purposes. remembering if theres less gens in their database then ite COI you get will be more favourable than is possibly reality...

[kelco]

A question-----

If mate select calculates the COI @ 25% and a the two dogs have no medical/genetic problems. Does that mean that the COI could be used as a guide towards potential inheretance of positive traits ?

[arwesa]

yes you have an equal chance of positive and negative doubling up..... thats the point of linebreeding. thats why you then do a complete outcross to introduce fresh dna and reduce the COI again... linebreeding in itself has to be a good thing, its when you end up with a dog from very few ancestors you get unmanageable inbreeding as southerncockers pointed out on the mateselect thread.

but e.g. you can linebreed a permissable close relation and still get a relatively low COI e.g. if what was behind the rest of the pedigree was different.

Might be worth naming some past 'stallions' of the breed and seeing what their COI's were.....

Clarburgh Art ----- 5.3%

Kenine Robb of Rytex ....5.4%

Kettlestang Comet .....6.4%

current breed ave ....8.7%

any more?

[southern cockers]

Arwesa,
To continue on our previous discussion and what I inherently believe to be a flawed reading ( incidentally anytime % is used it surely relates to a rating out of 100).
I have just input Argyll Warrior C.O.I. and have received an index reading of 6.1% based on 5 complete generations over 11 gens computed.
I have then input a bitch of mine and received an index of 16.2% over 7 gens of 12 computed.
When I do a fictitious trial mating I receive a reading of 12.1% over 6 gens of 13 computed !
If the system can compute a fictitious mating over 6 gens why can it not compute a singular 6 gen mating for Argyll Warrior in the first instance ?

[scot-ish]

mateselect hasnt invented anything new here, its just opened up to all an easy way to get to as accurate a COI and all historical health info as is available at present for a given dog or mating.

Let me state at this juncture that I have long been a proponent of and derived immense satisfaction from the research of pedigrees using Wright's C.O.I. and the associated ancestral % loss long before the K.C decided to adopt such a programme and as such I welcome their efforts .
However it is absolutely INCORRECT to state that the readings from the K.C Mate Select programme are in any way ACCURATE as like is not being compared with like. In fact they are quite misleading save for singular dogs C.O.I.
If someone can explain to me that there is no discernable difference in the calculations of an index derived over 9 generations as applies to some readings compared to 15 generations that applies to others I will take the next plane to Mars !
I welcome the programme and its associated benefits in assessing the virtues of any potential mating but until such time as all dogs are equally assessed and a common denominator of generations applied in each instance the index/ coefficient is virtually worthless !

If you go re read the mateselect thread. I took the time to research the way COI's are calculated, the algorithm the KC are using and what happens when it doesnt have the data for past breedings. fact is the more gens you include in a COI the more accurate it is...its a guide, so using a predetermined cut off of 10 gens and then assuming theres no inbreeding prior to that (as you would with your software) is even more misleading...

As I understand it, the COI is a guide to tell you with as much accuracy as possible how much inbreeding there is for a particular dog its not a score out of 100 based on the last x generations. ideally to get a tru COI you need all the breeding info for a dog from the day the breed existed but obviously thats just not possible, so you use as much info as you can and have to assume where you dont have info prior to that there was no inbreeding (i.e. assume the best)...not ideal but thats the way it is. giving you how many gens they are going back and whats the shortest number of gens they have on record just gives your more discerning user a guide to how accurate the score is vs another dogs COI for comparitive purposes. remembering if theres less gens in their database then ite COI you get will be more favourable than is possibly reality...

so with your now not backing up why its so good. you stated in the other thread, you had a number of completley different sires, which you would use for your bitch, all of which, would probably put something different into the bitch, at least prosepectively.

for example.

youve stated, you would have. for example 3 sires you like.

sire 1 = COI 15% 9 generations
Sire 2 -=COI 17% 10 generations
sire 3 = COI 19% 12 generations.

based on this, which sire would you choose? bearing in mind the KC dont tell you how many generations its taken over. so they have given you 3 COI, and you would now pick the lowest, but these 3 COI's are not really directly comparable to each other.

for this i believe this tool is not worth its hassle, and will only confuse others more.

[arwesa]

why would it need to.

Argyle warriors past is obviously not complete in the database for more than 5 gens, and the longest it goes back is 11 gens....anything outside this will be considered a 0 COI...i.e. no interbreeding

your bitch they have 7 full gens on file and some parts go back 12 gens again anything outside this is counted 0%

when you do a mating the pups become a new generation so for them the system will have full history of 6 gens (theirs plus shortest link is Argyle warrior) and longest line info is your bitch which will now be 13 gens back....

nothing about it is fictitious...it gives as accurate a COI as is possible, using real data where it is available and assuming best case where it isnt....over time the COI's will get more and more accurate.

I really cant see your point.

[arwesa]

so with your now not backing up why its so good. you stated in the other thread, you had a number of completley different sires, which you would use for your bitch, all of which, would probably put something different into the bitch, at least prosepectively.

for example.

youve stated, you would have. for example 3 sires you like.

sire 1 = COI 15% 9 generations
Sire 2 -=COI 17% 10 generations
sire 3 = COI 19% 12 generations.

based on this, which sire would you choose? bearing in mind the KC dont tell you how many generations its taken over. so they have given you 3 COI, and you would now pick the lowest, but these 3 COI's are not really directly comparable to each other.

for this i believe this tool is not worth its hassle, and will only confuse others more.

On the other thread I just gave examples using sires that had previously been discussed on the forum on a thread about top studs.....just to give ideas of the types of numbers people would get.


I cant see how you think Im saying this isnt useful?

how can it confuse? like all things you know the more data thats used the more accurate result you get? how many gens is a red herring, the more gens the more accurate the COI. as you go back more gens the number of matings in the calculation increases exponentially and the effect an individual mating has on the overall COI diminishes...its unlikely theyre 0 which is why more gens is better to make it more accurate.

[scot-ish]

i think it will confuse, because the kc have made it public like this, and people will just assume the COI is very important, and probably the most important thing.

i doubt i will use it much, if only for a nosey at COI, but two dogs, one giving pups of a COI of 12, and the other 20. i doubt it would make a difference to my plans of which one i had favoured in the first place.