|
Development Forum |
 |
| Question about extra pigment genes | |
|  For Science!
InsanityPrelude
 
|
6/20/2013 | |
So I sat down to apply the CFE genes to a couple more breeds that needed them, and I noticed the Undine Norns have a buttload of additional pigment genes at the bottom of the genome- also the original ones are gone, which basically moves everything else up several slots.
This caused me a little concern, because as far as I can tell from running the description through Google translate, the extended color genes in Amanora's breeds caused problems with slider offspring.
I'm looking at the "fixed" Floras, and they have the same pigment genes as Chichis etc. but at the bottom of the genome- so from this, can I take away that it doesn't matter where the pigment genes are, as long as any extras are at the bottom? |
 Sanely Insane
RisenAngel
     Manager

|
6/20/2013 | |
I think it does somewhat matter where the pigment/pigment bleed genomes are in a genome. In addition to the crossbreeding troubles with Amanora's breeds, the Colortrue creatures have their pigment/pigment bleed genes scattered to the four winds within their genome and also tend to not crossbreed well.
On the other hand, my breeds have the same extra pigment/pigment bleed genes as Amanora's breeds do, but they're moved up so that they're lumped in with the default pigment/pigment bleed genes and I've yet to run into a problem with crossbreeding them with the default breeds.
I have reason to believe that leaving genes at the bottom of a genome might actually be worse for crossbreeding, particularly if the genes above them have been moved around/deleted, but ultimately I'm not quite sure what effects extra pigment/pigment bleed genes might have on crossbreeding.
~ The Realm ~
Risen Angel's Creatures Blog
|

Malkin
     Manager

|
6/20/2013 | 1 |
When the game creates a baby genome out of two parent genomes, it flips two sets of cards together. If both creatures have their genomes organised in the same way, then the baby will also inherit a full genome - but if they aren't, then the baby could end up with half a brain or something like that.
My TCR Norns |

Feddlefew
  
|
6/20/2013 | 2 |
I ran a test awhile ago- When the breeding subroutine crosses two genomes, it seems to line up the genomes by gen number, starting with gene 1. So if genes have been moved, there's a good chance that the offspring will have duplicates and missing genes associated with each crossover point.
Example: Let's say we have a norn with the following genome
1.A-2.B-3.Ca-4.Cb-5.Cc-6.D-7.E-8.F-9.G-10.H-11.I-12.J
Where the numbers are the gene order and the letters are the genes. The C's here are pigment genes.
Our second norn has its pigment genes evenly spaced:
1.A-2.B-3.Ca-4.D-5.E-6.Cb-7.F-8.G-9.Cc-10.H-11.I-12.J
If we breed this norn with the first, and the crossover points are between 3-4 and 8-9, then the child genomes are:
1.A-2.B-3.Ca-4.D-5.E-6Cb-7.F-8.G-9.G-10.H-11.I-12.J
and
1.A-2.B-3.Ca-4.Cb-5.Cc-6.D-7.E-8.F-9.Cc-10.H-11.I-12.J
While the first child is only missing a pigment gene, the second child is missing G! Hope it wasn't anything important.
Moving all the pigment genes to the end produces creates this genome:
1.A-2.B-3.D-4.E-5.F-6.G-7.H-8.I-9.J-10.Ca-11.Cb-12.Ca
With the results of the same cross as above being:
1.A-2.B-3.Ca-4.E-5.F-6.G-7.H-8.I-9.G-10.H-11.I-12.J
and
1.A-2.B-3.D-4.Cb-5.Cc-6.D-7.E-8.F-9.J-10.Ca-11.Cb-12.Ca
The first is missing pigment genes, while the second is missing G and H.
---
So moving all the pigment genes to the end does cause more problems then scattering them, because it increase the probability that a block of genes will be missing.
|
 For Science!
InsanityPrelude
 
|
6/20/2013 | |
So that means the "safest" thing would be to move some of the pigment genes back to where they are in the standard genomes. A bit tedious, but easy enough. ![[nsmile] [nsmile]](/images/smilies/emot_smile.gif) |
 Code Monkey
evolnemesis
    
|
2/21/2014 | |
Colortrue breeds having those extra pigment genes and scattering pigments throughout the genome might mean you get an 'unsafe' genome for crossbreeding, with a fair chance for bad mutations if you breed them with breeds that aren't colortrue, but you get the same thing with CFE genomes as well... Also, having colortrue genetics has many positive effects too, and fixes some annoying problems with the normal, blocked-together pigment genes.
Here's the idea: Since the genes of two parents are lined up and spliced together when mated, with the offspring getting dna in chunks from one parent or the other between crossover points, the number of pigment genes and pigment scattering in the colortrue norns makes it very likely that the offspring of colortrues will have a good mixture of pigment genes from each parent.
Because each group of pigment values in a norn are averaged to get the final color (the groups being reds, greens, blues, rotations, and swaps), colortrue offspring have a very cool effect of being a color that is a nice mix somewhere between the colors of the 2 parents (i.e. A red colortrue + a blue colortrue will give you a violet/magenta/purple color, a green+blue gives you a turquoise/cyan/aquamarine color, a red+purple give you a magenta/mauve/mulberry color, etc...)
The extra pigment genes and pigment spreading in the colortrues also mean that whole pigment groups are a lot less likely to be entirely lost to random mutation or slight sliding effects from one parent having a few more genes; something that can cause drastic and usually undesirable effects in your color (for example, if you have a nice pink norn and you were to lose the whole red group, the result would just look washed out and too bright with no special color at all)
On the other hand, if all the pigment genes are blocked together, like in most normal C3/DS genomes, then you get the tendency for most or all of them to just be carried all together from one parent when breeding happens... This means that rather than a mix, you tend to get offspring that only favor one parent's coloration, with only a small minority or even no pigment genes from the other parent.
Even when there IS a crossover somewhere in the blocked pigment genes, the crossover will tend to mostly affect only one group, like the reds, greens, or blues, making the part copied over to the offspring just a tendency to a partial shade that looks nothing like EITHER parent's color, and even that bit still gets averaged out and muted by the other parent's pigment genes.
So, rather than ever getting any kind of actual mix of the 2 block-pigmented parents, you will instead see something looking just like one parent or the other; only occasionally with a randomish shift in the red, blue, or green part of their pigment. These occasional shifting effects of partial pigments are pretty much the same effects that have a decent chance to appear randomly with mutations anyway even if you have no crossovers or have 2 parents with the same exact colors.
This also means that if either block-pigmented parent has some pigment genes you really like, or a cool mutation in some part of their pigment or color bleed, you have a good chance (over 50%...) to either have ALL those pigment genes you want totally thrown out, or to lose whole major groups of them that are necessary to make them ever show up properly in future generations.
"For small creatures such as we, the vastness is bearable only through love."
"We are a way for the cosmos to know itself." - Carl Sagan |

Missmysterics
 
|
6/22/2014 | |
So the best bet for extra colour genes is to put them with the rest of the colour genes? |
 Code Monkey
evolnemesis
    
|
6/22/2014 | |
If you want them to be able to crossbreed with other creatures easily without a good chance of sliders, the best bet is just to put the new color genes at the end. This also separates the new group of pigment genes from the normal pigment gene group, so it's a much better possibility that creatures will inherit just some color genes from each parent.
"For small creatures such as we, the vastness is bearable only through love."
"We are a way for the cosmos to know itself." - Carl Sagan |
|