I'm a bit confused regarding some aspects of breeding for deviance. My understanding is that a homozygous base pair grants a 10% chance of deviance, but what if both skin color and skin pattern are homozygous? Is there a 10% chance of deviance per base pair, or is it still just a 10% chance that one of them will be deviant? In either case, how does the system decide which deviance will occur?

I'm trying to figure out if having double homozygotes will increase my probability of yielding deviants, and if so, by how much. Whatever is going on here definitely isn't Mendellian genetics, so I'm trying to get an idea of how it works. 

I believe it is always 10% chance given the correct circumstances. 

From news update(http://beastkeeper.com/thread/5473): 

You need to select parents that carry same color or/and pattern genes. A baby born as homozygous (carrying two copies of the same gene) for color or pattern has a 1/10 chance to develop a deviant color when maturing. This mutation is phenotypic and not inherited by the offsprings of an affected pet. A gene that produces a deviant phenotype will be marked with a star.
The color name of such pet gets an asterisk (*) symbol to mark the deviation.

The only way to breed a pet with both deviant color and pattern is to choose very specific parents. One of the parents should have deviant color and the second one should have deviant pattern. The chances of getting a deviation on each color parameter still remains 1/10 so the babies with double deviation are extremely rare and valuable.

Let's look at some DD pets: 

http://beastkeeper.com/pet/295514

This pet is DD for White and Striped.

Father is White/Grey and Striped, and the Mother is Chestnut and striped/cream.  So in this situation the parents weren't both doubly dominant for white and striped, or even the same deviance between them! But both parents were deviants. 

This pet is DD for White/Leopard

The Father is white/grey and striped, and the mother is grey and leopard/striped. 

So ths same case... the original deviant parents didn't necessarily be the same with the other parent, or even homozygous for the resulting offspring. 

From these two cases (small sample I know). I think the requirements have to have a parent that has a devious pattern, and one with a devious color. The chance of then having a deviant from deviant parentage is 10%. 

I think where it gets weird is how this merges with our own unique genetic system here. I think Timain gives a nice description of that here: http://beastkeeper.com/thread/5218/1#43165

However... when breeding regular 'ol deviants I like to choose pets that are doubly homozygous, and the same as each other to increase the liklihood that two genes I want will overlap and create the pet I want, but the chance of breeding an actual deviant is always 10%.With double deviants any combination will do, they don't need to be homozygous. 

I think I may have made myself more confused, while strangly understanding more? Is that even possible?

Double deviance is a whole new level of crazy. That's definitely going beyond Mendellian laws by allowing phenotypic traits to affect offspring. But what I'm more concerned about is whether double homozygote parents have a 10 or 20 percent chance of yielding deviant offspring.

For example, if both parents are the same homozygous color and pattern, is there still just a 10 percent chance that the offspring will be deviant, or a 10 percent chance per homozygous base pair?

The wording in the description seems to hint at the possibility of independent calculation. Up top, it says "a baby born as homozygous for color OR pattern has a 1/10 chance to develop a deviant color while maturing", implying that the 10 percent chance only applies to one or the other, and further down, it says "the chances of getting a deviation on EACH color parameter still remain 1/10", likewise implying that they are calculated independently.

This is important for me since I'm breeding for deviants and need to know if double homozygous parents are worth the extra trouble. If it's a 10 percent chance either way, I don't need to waste my time looking for matching doubles, but if it doubles the chance of deviant offspring, it's worth finding or breeding them. 

Oh I see now!! Hmm. In that case I'm not quite sure of it. I think it defies what you would think of how genes typically behave, and is a flat rate of 10% regardless of all other factors. When I was looking through the pets I saw trying to describe the process it seemed that non-homozygous pets easily created deviants. However, I can't accuratly say this because I don't know how many other pets were bred and were faulures. I don't have a large sample size. 

I see where the confusion arises. I'm not trying to breed deviant parents for double-deviant offspring. Instead, I'm trying to breed normal parents for deviant offspring. To the best of my knowledge, this is how it works when there's a single homozygous base pair involved:

Let's say I'm breeding for red deviants. There are three potential probabilities, based on how homozygous the parents are.

LOW PROBABILITY
Father: Red/Blue Mother: Green/Red
Chance of Red/Red: 25%
Chance of Deviant on Red/Red: 10%
Overall Chance of Deviant: 2.5%

MODERATE PROBABILTY
Father: Red/Blue Mother Red/Red
Chance of Red/Red: 50%
Chance of Deviant on Red/Red: 10%
Overall Chance of Deviant: 5%

HIGH PROBABILITY
Father: Red/Red Mother: Red/Red
Chance of Red/Red: 100%
Chance of Deviant on Red/Red: 10%
Overall Chance of Deviant: 10%

My question is, what happens when both color AND pattern are homozygous? Which of these calculations occurs?

INDEPENDENT PROBABILITY
Father: Red/Red & Spotted/Spotted Mother Red/Red & Spotted/Spotted
Chance of Red/Red: 100%
Chance of Spotted/Spotted: 100%
Chance of Deviant on Red/Red: 10%
Chance of Deviant on Spotted/Spotted: 10%
Overall Chance of Deviant: 20%

SHARED PROBABILITY
Father: Red/Red & Spotted/Spotted Mother Red/Red & Spotted/Spotted
Chance of Red/Red: 100%
Chance of Spotted/Spotted: 100%
Chance of Deviant on Red/Red OR Spotted/Spotted: 10%
Overall Chance of Deviant: 10%

And in either case, since it is stated clearly that the only way double deviants can occur is from deviant parents, how does the system decide which base pair becomes deviant? Is it randomly selected? Is it based on relative expression power like normal genes?

Senorali, we're using a lot things that aren't Mendellian, like multiple allelism and complementation :)

As for the game mechanics, the chance to get a deviant offspring is always 10%. First we're rolling the chance if the child is deviant and then we're randomly rolling what trait exactly is deviant. Double homozygous parents will have the same number deviant offsprings with single homozygous ones. The sentence about "still 10% for each trait" supposed to mean that chance to get a double deviant is 1/10 * 1/10= 1/100. I apologize if the wording isn't clear enough. English isn't my native language and we don't have a native or a С1/С2 level English writer to help us, so we're relying on our players when it comes to wording.

If you're interested in background, here it is. Deviant zygote is a pathology, it contains a complex of abnormal proteins that is recognized as an alien or ill cell and usually destroyed by the mother's immunity if it recognizes the pathologic marker sequences. We consider that due to a number of different causes this factor isn't interpreted as an alien and immunity fails sometimes, and the embryo develops successfully. This is possible only with one deviant gene for a normal animal while two and more will always be destroyed. Deviation factor present in mother organism supposed to let the zygote with double deviant mutation live while in regular organism it will be desroyed by the antibodies. This is somewhat similar with how Rh factor works in humans including those exclusions when Rh- mother successfully bears Rh+ child.

Thank you for the thorough explanation! 

That makes much more sense now. It is independant. Thank you Angel!