Het percentages??

[Rob @ RK Reptiles]

I am curious to why people advertise an animal as a 25% or 33% het as there is no such thing. It is either a 100% het or a 50% het. Breeding a possible het to a normal makes those babies normals until the parent is proven out. Any thoughts on this everyone?

 

[Seamus Haley]

I am curious to why people advertise an animal as a 25% or 33% het as there is no such thing.

Sure there is Rob.

First off, as you well know, an animal either IS het or ISN'T, but we haven't identified genetic markers and the only way to test the hetness of an animal is to breed it out or produce it in such a fashion as to leave no doubt.

But when you take possible hets and breed them to other possible hets or normals, you can get any percentage you want given enough repeated applications.

I'm going to assume everyone knows how to get a known het, a fifty percent possible and a sixty six percent possible het...

Now... cross a possible het with a normal, a known het or another possible het. It requires multiple punnet squares proportional to the chances that either parent animal is truly heterozygous or not, as it's unknown at the time of copulation (unless you start dropping offspring which are homozygous for the recessive trait, at which poiint everything gets revised since both parents are known hets).

Given a streak of poor luck or a species with small cluch numbers, you could easily go three or four generations without producing an animal displaying the recessive phenotype but with offspring carrying the gene. Of course as further evidence piles up and the animals continually fail to produce any, the "possible het" status becomes less and less likely... and even given pure mathematical mendelian probabilities, a twenty five percent possible het is not a really good buy if it costs any more than a normal animal, but can be an additional selling point of minor interest.

 

[CornNut]

I like the point about there being no real such thing as a 50% or 66% het either because each individual is either a het or not a het. So if you are going to use the concept of a 50% or 66% possible het to account for the probability then why not continue the logic on to 25% and 33%?

I do think it is important for the buyer to realize the all or nothing nature of this extra generation removal from a for sure het. That is, it doesn't mater how many 25% chance hets you buy out of the same 50% chance het male if he doesn't hit his 50% chance. But if you where to buy a large number or 25% chance hets out of a large number of unproven 50% chance hets you could expect 25% of them to prove. Some people have trouble with the concepts so the burden would be on the seller to properly explain 25% and 33% chance hets.

Another angle on 25% chance hets is if you have any 25% chance het stripe females you want to sell me at normal hatchling price just let me know. I don't sell any of my 25% chance het females; it would be hard to know what to price them at if I did. I do sell my 25% chance het males at normal prices but I do tell the buyer what they have a chance to be het for in case they want to use them to start a project and upgrade to a known het male the next generation. Some people actually breed normal males for budget reasons so with 25% chance hets at the same price they could at least have a shot at getting started.

 

[Seamus Haley]

Some people actually breed normal males for budget reasons so with 25% chance hets at the same price they could at least have a shot at getting started.

The converse of that of course being people who do not desire any possibility of a certain genotype being involved in their breeding project. If someone is working towards a given result, they would want to avoid even low percentage possible hets which are not conducive to the animals they wish to produce, so advertising even unlikely possible hets AS possible hets is important for an honest dealer.

 

[Mark and Aimee]

So if you are going to use the concept of a 50% or 66% possible het to account for the probability then why not continue the logic on to 25% and 33%?

We all know that it takes a KNOWN het to produce a 50% or 66% het. The percentages are such because we KNOW half the clutch of 50% hets should actually be hets. And we KNOW 2 out of 3 of the 66% hets should actually be hets.

But - with your 25% hets and 33% hets, you're just taking a wild guess. There's a fair possibility that it is just a clutch of normals. You don't have a definite gene there to start with. That's the big difference.

 

[CornNut]

I'll agree that the 25% and 33% chance hets are further removed and this is an important difference that I hope everyone understands. Given the general level of confusion on genetics and statistics it is something that should probably be explained in detail every time 25% and 33% hets are offered.

However, with the 50% and 66% hets you are still just guessing! Just look at all the pastel X normal and spider X normal clutches. Before the eggs hatch they are 50% chance hets for pastel or spider. Once they hatch you see that you don't always get the 50/50 split in the clutch. The misinformation that 50% of a clutch or 66% of the non-homozygous in a clutch will be hets is just a guess also. In the long run it works out to be a good guess just like 25% and 33% hets from a large number of 50% and 66% hets respectively works out to be a good guess. However in an individual clutch it is just a guess and not a guarantee.

 

[Seamus Haley]

We all know that it takes a KNOWN het to produce a 50% or 66% het.

Technically by crossing and recrossing possible hets, hets and animals which are homozygous for the recessive trait in question you could get any percentage you wanted from zero to one hundred, it's just that after a few repeated breedings most animals which actually ARE heterozygous will have proven themselves out, thus changing up all the numbers for all of their offspring.

But if we could assume the worst luck in the world when it comes to producing animals which are homozygous for a recessive trait, given enough generations and combinations, you can technically get those percentage numbers anyplace you want them.

The percentages are such because we KNOW half the clutch of 50% hets should actually be hets. And we KNOW 2 out of 3 of the 66% hets should actually be hets.

Sorry to have to put it like this but... That is dead wrong. The possible outcome proportions produced from punnet squares represent the individual chances for each seperate zygote, each egg, each neonate. Start flipping coins or rolling dice, within a small number (say the average clutch numbers for popular species) you *probably* won't get results that are totally statistically accurate. The chances are reset each time.

Since 25% and 33% possible hets are only a clutch removed from 50% or 66% possible hets and given the clutch numbers for most species, it's entirely possible and the production and labeling of such animals is just as valid as labeling possible hets with a higher percentile chance.

You don't have a definite gene there to start with. That's the big difference.

I can agree with that though and see your point... you're saying essentially that it's emotionally a sucker's game because there was no animal with concrete genetics involved directly in the production. This is true, but statistically such possible hets ARE accurately represented.

And CornNut... Excellent post.