Possible Het Input to Morph Calculator [#1189]

perfect explanation. if this person truly wanted a result of some kind the only thing they would be are poss hets at best.

1 Like

In the case that you have a 50% or 66% het, it’s a 50% or 66% chance that you get the results from putting the 100% het into the calculator. And a corresponding 50% or 33% chance that you get the results from not putting het into the calculator.

For example, if you breed a 66% het to a normal, the offspring would technically be a 33% het. Most breeders don’t go into this level of detail though, and this is how recessive genes get lost and then pop up in people’s collections.

Calculating the odds for other pairings get a little more involved, but it’s not terribly difficult to do. How inclined are you to learn a little probabilistic math?

2 Likes

All you have to do to know your outcomes is run both scenarios, first scenario considering that the animals are hets, second considering that the animals are not, because they either are or they are not.

5 Likes

Makes sense, thanks for the info

This is a feature request which is not that uncommon. I have renamed it to what we call it: " “Possible Het Input to Morph Calculator”, and moved it into the correct subcategory.

1 Like

No, they would not.

There is no such thing as a 33% poss het or a 25% poss het. The original 66% poss or 50% poss is either a het or it is not but you have to prove it out. The percentage is the statistical odds of the gene having been passed on from the breeding that produced it but those statistical odds begin and end for that one breeding only. The ultimate het status is still absolutely binary. As such, all subsequent breedings would fall under that same binary - If the odds were against you and so your 66% is not het, then breeding it to a normal means the offspring from that breeding would be 0% het (because half of zero is zero) not 33% poss het

3 Likes

Travis, I’m curious about this. You’re making this argument based on the conventions of what is understood by “pos het”, not mathematically, right? Because mathematically, and I could totally be mistaken, but 33% pos het makes sense for (66% x normal). I think you are just saying, it is not conventional for “pos het” to refer to subsequent generations.

5 Likes

No, because it isn’t established that the “66% het” is or isn’t a het, so you would be essentially taking the probability of a probability.

Imagine I hand you a box full of blocks, and there was a 66% chance there was a round ball in there. There might be, there might not be, you don’t know. Then some of that box got dumped into another box of blocks and you were told to estimate the probability that a ball was in that second box. If there was no ball in the 1st box, then there’s obviously no chance of it being the in the second box. Since we don’t know whether there was a ball in the first box, it’s misleading and untrue to say that there’s a 33% probability of a ball being in the second box

6 Likes

And sadly it has been done and seen in the herp industry

1 Like

I do believe in most cases it’s due to misunderstandings of how genetics and probability work rather than attempting to scam or mislead people intentionally, however it’s a definite problem. The concepts of 66% and 50% hets are very confusing to many and I’m not sure the best way to handle it.

Based on nothing but the simple mathematics you are correct, 33 is half of 66. However, what we are dealing with is probability statistics which are a whole different monster. Chester’s analogy sums it up pretty well (though it is not quite as ideal as I would like LOL)

1 Like

Respectfully, I don’t find chester’s analogy helpful at all. :). Outside of some predefined conventions about what “pos het” means (in a ball python sense), I would not have arrived at these conclusions. Taking probabilities of probabilities is exactly what you do in math. It’s not misleading or untrue to say that there’s a 33% chance, unless there is an additional layer of meaning put on that, which I’m calling conventions.

Simply put, the argument should go, when someone says “pos het”, it implies that those odds were calculated based on parents whose genetics were fully known. This is because it’s the convention in our community, not because it couldn’t be generalized.

Now I think an interesting side issue is that pos het is most accurate when the clutch is laid. Once other breedings have occurred, the breeder has accumulated evidence which is not factored into that label. If I have a 66% het, and I breed it 10 times and get no offspring, I now have good evidence it’s actually not het, but I could still call it 66% pos het technically. However this is true and an issue not just for the offspring (pos het of pos het, like what we are talking about) but also the first generation of pos het too. So while this is interesting, it’s beside the point.

5 Likes

Let me try this analogy then…

I have one red cube and one green cube in a box. In a pitch black room, I reach in blindly and pull out one block and put it in a new box.

I bring those two boxes out to you. I tell you your goal is to have a green cube at the end of this. You choose one of those boxes.

There is a 50% chance the box you picked is holding the green cube.

Now, without you opening it, I take your box away from you and go back into the pitch black room. I turn on the light, open the box, take the cube out. Turns out you actually picked the red cube. I put your cube in a new box. I also grab a clear cube and put it in a new box. I bring both of those boxes out to you and ask you to pick one at random and remind you again that your goal is to have a green cube at the end of this.

Now, I ask the question: With two boxes sitting in front of you, neither of which contain a green cube, what are the odds of picking a box holding a green cube? Is it a 25% chance? Or is it a 0% chance?
.
.

Well yeah, I guess you could do that. But if you did, that would make you a real jerkwad LOL

6 Likes

I’d argue that there’s possibility for hets ranging from 0% to 100%.

The ways to get above 66% het are kind of silly from a practical breeding perspective though. One example I can think of would require pairing poss het to poss het and getting a very small clutch with no visual offspring, then breeding that offspring and getting a visual result. There’s probably some other creative ways of making observations to update the probabilities upwards but not to 100% certainty of its genetics.

Probabilities less than 66% are possible both through multiple generational breeding of possible hets (which reduces the odds for offspring), and through observations of offspring from possible hets in which visuals are possible, but no visuals are hatched (which reduces the odds for offspring and parents). The reduction of the odds are finite, calculable, and never reduce the poss het to zero. Only to approach zero as a limit as you approach an infinite number of generations or observations.

1 Like

I’ve done a lot of thinking and I’m ready to chime in. Obviously this is a much debated topic and everyone on both sides has very solid input. I honestly believe everyone is correct. Now I’m not an expert on genetics, I’m not an expert on reptiles and I’m not an expert on ball pythons. I got my first snake over a month ago.

However, I’ve read a ton since and I’m familiar with basic genetics of other animals. And I do have the basic concept of how the DNA chain works and how certain genomes if present will fit or be present in certain spaces of that chain. My understanding is this is how you can have two supers in the same animal, one super morph sitting in one spot on the DNA chain and the other super sitting in another spot. Some morphs possibly take up the exact same spot on the chain so you have either one or the other, not both.

Now I see a lot of snakes that have the same listed morph make up but have subtle differences visually. This could be different lines of those morphs, this could be different underlying morphs working there way out. And any of these characteristics may not be able to be “proven” out or passed on. But I argue that a lot of times a certain unrelated combo may look slightly different from one animal to the next.

That’s why I was curious about being able to add percentages of hets in the calculator. I believe some hets fit in different spots on the DNA chain and even though are not fully reproduced as visuals, or be passed on fully to offspring, can still have an effect on the outcome. If you end up with two unrelated products that are, for example, both Pastel. One is Pastel 0% het, the other say Pastel 10% enchi. I bet those two side by side may be noticeably different, maybe not. If you kept repeating that breeding, I bet you would come across a noticeably different snake, what? 1 out of 10

First of all there is no het Enchi, Enchi is a codominant morph. Second of all, as we have said, being a het is binary. A snake either is or isn’t het for a recessive trait. If they have two copies of the mutation/SNP/genetic rearrangement at the locus for that phenotype, it’s a visual morph. If there’s one copy, it’s a het, which can have some some visual cues, such as a ringer with a het pied morph. A “10% het” has a 10% change of being a het, and it might have visual cues relating to whether it’s het or not, but if you kept breeding it, you’d continually dilute the chances of getting a het out of that pairing and you would absolutely not come across a noticeably different snake. Because that’s not how it works at all

4 Likes

So the confusion here might be that the notation 50% het refers to the probability - the chance - that a snake has inherited a gene. Think of it as a genetic coin flip from breeding the parent.

With genes like Enchi, which are incomplete dominant (called often called co-dominant in the hobby), you can see the result of that coin flip: it looks like an Enchi. So you can call a heads a heads and a tails a tails.

With recessive genes, e.g. pied, if the snake with a het pied parent comes out looking not-pied. Now you can’t see the result of that genetic coin flip. It might have het pied, it might not; because both of those results will look exactly the same. Think of it like you’ve done the coin flip but you’ve still got your hand over the coin. How can you describe the state of that hidden coin? The best you can do is say there’s a 50% chance it’s heads, 50% chance it’s tails.

But similar to the coin flip, it can’t be a little bit heads plus a little bit tails under your hand. It can’t be a little bit Enchi plus a little bit normal; nor a little bit het Pied plus a little bit not het Pied. The reason is that what are called “morphs” in the Ball Python hobby are all single-gene mutations, not the result of many genes mixed together that can be inherited proportionally.

Hope that helps clarify a little bit.

I do not understand how you can make this comment…

And still believe this statement as true.

These are two mutually incompatible statements.

You are correct that it is a coin flip, and that coin flip determines the probability. The probability of each flip of the coin is wholly and completely independent of any previous flip. It is not a reductive process. If I flip a coin I have a 50/50 chance of it landing on heads. If I flip it a second time, I have the same 50/50 chance of getting heads, not a 25/75 chance.
.

Running the analogy to breeding poss hets. I breed a het to a normal. Those offspring are 100% het. They are a normal coin with a head side and a tail side. I buy one of those and breed it to a normal, which is a trick coin that is tails on both sides. I flip both coins and then I glue them together such that whatever faces land upright become the new faces of the coin.

I do not even have to flip the WT coin because it is tails regardless. I flip the het coin and cover it before you can see what it is. There is a 50/50 chance it landed heads up. You roll that risk and buy the 50% possible het from me. But I peaked and saw that it was tails up. The animal is, in absolute fact a 0% het. It is now a trick coin that is tails on both sides.

This animal does not carry the het gene. As such the, probability of any offspring produced from this animal ever inheriting the het gene from it is ZERO. Therefore, all of its offspring are 0% probable het.

If you say it is a 25% probable het, that implies there is a 25% chance that any animal you chose from the clutch could carry the het gene. It is a dishonest lie to call it a 25% het because (random spontaneous mutation aside) there is not a single chance you will ever ever get the het gene. The parent is a trick coin and will always always always pass on the tails/non-het allele

2 Likes

To go along with how I’m understanding your “trick” coin analogy: 2 consecutive coin flips. One with a heads/tails coin. The second flip is with either another heads/tails coin or with a tails/tails coin. If the first flip is heads, we get the fair coin, if the first flip is tails we get the “trick” coin. Some honest referee is flipping the coins as prescribed, but we can’t see the results of the flips.

The first flip is our 100% het bred to a normal. The second is the 50% pos het to another normal.

At the end of the second flip in the game above, what is the probability that the result of that flip is heads? Is that impossible to quantify?

For that second flip the conditional probability of Heads given the second coin is the “trick” coin is 0%
Also, for the second flip the conditional probability of Heads given the fair coin is 50%
But we are interested simply in “what is the probability the second flip is heads.”

We don’t know the exact condition we are in after the first flip though, so we can’t rely completely on the conditional probabilities. But we do know the probability for being in each of the conditions for those conditional probabilities: in this game it’s 50/50.

The way the math works out for this:
probability of H2 = (probability of H2 given H1) * (probability of H1) + (probability of H2 given T1) * (probability of T1) = 0.5*0.5 + 0*0.5 = 0.25

So yes, breeding a 50% het to a normal means that every hatchling in that clutch has a 25% chance of having inherited the het gene. Further observations of genetic interactions can update this belief for all the animals involved. But as it stands after that pairing: it is certainly honest, theoretically, and mathematically correct to state that there is a 25% probability, chance, odds, that those snakes have the gene. *assuming we can ignore the probabilities for other genetic mechanisms; which I am certainly no expert on

1 Like

I have a genuine question because I have been reading through this and following but am still SO confused.

Breaking this down to basics: If I have a Normal female and a het pied male. I breed them. Their offspring will be 50% het pied, right? So 2/4 babies will be het pied, statistically.

What if I never prove out the het pied babies and breed one to a Hypo a few years later? Offspring from that clutch would then be 100% het Hypo, “Pos” het pied, right? Then later on next season, if I prove out the 50% het pied parent from the first clutch as 100% het Pied, the baby would then become 100% het hypo, 50% het pied, right?

My question is what % would the “pos” het baby be before being proven? Just keep it “pos” het or call it 33% het Pied?

I’m just trying to put this in a way I could understand better. haha

1 Like