Real live genetics

[bamboe32]

Real life genetics

What is the real average outcome when crossing double hets.
For instance crossing het amel & het aner with het amel & het aner.

The calculation says you will get 1/16 snow.
Meaning if you have 16 eggs you will get 1 snow.

But I am sure nature wil not look at those calculations and you will
get more or less snows.

What is the average everybody gets, so I know what to expect in real life
when crossing colors.

Rob.

 

[Menhir]

Hmmm - these are real life ratios.
But remember, having a chance of 1/16 does not mean, that you have one in every 16 eggs.
Take a dice, the chances for 6 is 1/6 - but that doesn't mean you have one 6 every 6 times you dice.
The chances not to get a 6 are (5/6)^6 and that are none the less about 33%, so in one of 3 cases you won't get any 6 of you dice 6 times. Do you know what I mean?

The 1/16 is for every egg and is "diced" everytime, for every egg new. The thing with the hatchlings is like dicing every egg with a 16 sided dice with beeing one side "snow". So, now you can imagine how different clutches can look like. BUT over thousands of clutches - you will have about 1/16 snows and that does mean, that you have 1 snow every 16 animals

 

[Jimmy Johnson]

ratios

These numbers are only to give you the mathmatical probability.
Nature can and does play by her own rules.
I bred a ghost het amel to a snow.
This should have yeilded 50% snows and 50% anerys.
Out of 16 eggs 10 were snow and 6 anery.
that would be 62.5 % snow and 37.5% anery.
So, with nature and corn snakes these numbers are not "guaranteed" but they do give you a reasonable idea of what to expect.

 

[Serpwidgets]

The calculation says you will get 1/16 snow.

Meaning if you have 16 eggs you will get 1 snow.

No, that is not at all what it means. It means that each egg has a 1 in 16 chance of being a snow. That is completely and totally different from saying "if you get 16 eggs you will get 1 snow."

If you have 16 eggs from double hets amel/anery:
There is a 35.6% chance of hatching NO snows.
There is a 38% chance of hatching 1 snow.
There is a 19% chance of hatching 2 snows.
There is a 5.9% chance of hatching 3 snows.
The chance of hatching 4 or more snows is about 1 in 66.

 

[Serpwidgets]

These numbers are only to give you the mathmatical probability.
Nature can and does play by her own rules.
I bred a ghost het amel to a snow.
This should have yeilded 50% snows and 50% anerys.
Out of 16 eggs 10 were snow and 6 anery.
that would be 62.5 % snow and 37.5% anery.
So, with nature and corn snakes these numbers are not "guaranteed" but they do give you a reasonable idea of what to expect.

I don't know where people get "should have" out of these predictions. There is only a 1 in 5 chance that a clutch of 16 will be split between 8 and 8.

In any clutch of 16 with two equal possibilities, there is a 1 in 4 chance of getting a 10/6 or 6/10 split. These together are more likely than getting what people will say "should" come out.

The chance of getting anything from and including 6/10 up to 10/6 is almost 80%.

If you have 1000:1 odds against something happening, that means that 1 in 1000 times the outcome will "defy the odds." We should expect a decent amount of variations in any clutch.

Also, the larger the clutch, the less likely your outcome is to exactly match the prediction, but the more likely it is to be very close to the prediction. The smaller the clutch, the more likely you are to have "way different from the prediction" results. :grin01:

 

[Jimmy Johnson]

nnnn

Like I stated, The percentages are just a guide (and in my opinion not a very good one) and definitely not guaranteed.

I definitely don't beleive they "should" be this or "should" be that.
I was actually being more sarcastic then anything. That is why I posted the results. an example of the "shoulds" not turning out.
I probably should have used the quotation marks to show the intent of my sarcasm.

 

[Serpwidgets]

People tend to take "50% per egg" and translate this into a nonsensical conclusion of "in a clutch of 16, you should get 8." This may be the most common translation, but it has nothing to do with reality. For anyone who does take the predictions this way, they are useless. That's their loss.

However, for those of us who understand what these predictions mean, they are very useful. A hammer is only as useful as the carpenter who uses it, and it is a bad carpenter who blames his hammer. If these predictions were useless, your computer, the one you're using to argue this point and read this post right now, would not exist.

The "prediction" based on a 50/50 chance per egg, looks like the following chart for a clutch of 16. As I mentioned before, getting a 6/10 split on a clutch of 16 is in no way defying the prediction, or anything out of the ordinary.

 

[Menhir]

Nice discrete gaussian distribution
I'm quite shure that I've argued a related thing with Jimmy already in the past times and that it seems thats a little like the genetics thing - if you don't have a good fundamental knowledge about probability calculation, you may run into wrong explanations very fast and it's quite hard to get out there.

One fundamental thing with the progencys is - FORGET THE WORD CLUTCH!
Clutch doesn't exist for these ratios - it's a perentage over every single animal and therefore only true for large NUMBERS of animals, NOT clutches.

If you examine 200 clutches with about 3200 animals, then you will find the ratios given. If you examine special numbers of eggs (such as clutch size) you have to calculate the propabilities for getting a special result new. As Serp and I stated above, that's not the same as the ratios for every single animal.
(or as I did with the dice, perhaps this can be proven by everyone at home)

 

[El Jefe]

Het Lava X Het Lava

Last year I did Het Lava to Het Lava....


Out of about 8 eggs that hatched...6 were lavas and 2 were normals....far different than the projected 1/4!

Oh well, I didn't mind!


But....that being said. I've seen the opposite happen as well. So, genetics are crazy and in REAL LIFE you never REALLY know what is going to happen. Sometimes you will 'beat' the odds and get more, sometimes less. All the ratios are a general prediction of what you SHOULD have if everything matched up like Gregor Mendel stated in the mid 1800's. Even Mendel, however, remarked that it must occur over time and that one generation may not show the exact prediction but over time one should see it even out.

The big idea and key words are: ON AVERAGE you should see....blah blah blah.

On average, over time, you will see the system become stable and reach a value of close to the predicted. One clutch has a lower chance of hitting that average than 10 or 100 clutches averaged together.

 

[countMEout]

With odds and percentages they are just a guide and one time you might get a yeild of 50% when you were expecting a 25% yeild. But over the long term when dealing with all of these genetics exchanges the expected average would tend to come closer to the expected average. So you cannot think because you got a favorable payout now the numbers are bunk it is all variable and the percents are a guideline that you would begin to see more overtime when involved in breeding. You cannot take 1 case as hard science as nice as that would be.