So is it more like there's some secondary "thing" that is for hair color and it looks at that gene to determine the color but there can also be another secondary "thing" such as hair length that also uses that same gene to determine that?
I think it may help to not think of genes as lookup tables.
It's rather that genes act as blueprints for compound (protein/RNA) factories (which can also be potentially turned off/on). So the compounds that is produced may interact with the compound(s) that end up resulting in the hair color, and it may also interact with the compound that results in the hair length.
The problem is that any compound may in theory interact with any other compound, and there currently exists no way to 100% determine that they won't (for most compounds), which leaves open a huge space of possible chains of interactions.
In practice there are many interactions for every biological compound, which is the reason why medicine is so hard to develop and usually has a risk of side-effects.
It's also worth remembering that a cell isn't a well structured environment - it's essentially a bag of chemicals, with some slightly smaller bags of other chemicals inside it.
While there's a lot of mechanisms which are adding order and structure to what happens, it's all still just a big concentrated aqueous solution of everything in the cell, diffusion processes and mixing and all.
So statements like a gene being "switched on" are very much an abstraction: whereas switching on a data line in a chip puts a very nice neat little voltage potential somewhere, switching on a gene basically just means the concentration of some "chemical" (protein) starts increasing and getting mixed into the cell (or ejected out of it by interacting with a bunch of other floating around things).
> While there's a lot of mechanisms which are adding order and structure to what happens, it's all still just a big concentrated aqueous solution of everything in the cell, diffusion processes and mixing and all.
Any closer look into biology completely dispels the notion of intelligent design. It’s over complicated, fragile, poorly architected and works, essentially, by side effects of all chemical reactions happening inside those bags of goo.
ARHGAP36[1] stimulates GTP catabolism. And since MC1R is a g protein coupled receptor and uses GTP as an energy source, changes in the levels GTP will change the response of MC1R[2] to MSH and ACTH. It will react less strongly when there is lower GTP, and more strongly when there is higher GTP[3].
A rough analogy is like if you flipped some bits in a computer program's executable code and then looked at what happens when you run that code, and notice a specific feature now no longer works for example. Yeah you found a way to break something, but you don't know what else you broke, or what the full effects were of flipping those bits.
Also, the results with the current methods could be fine for imprecise purposes. Say, if you had some cat embryos and wanted to know which one was orange for your designer kitten.
But in the future it might not be good enough for later more precise interventions; if you just start editing cats to make them orange now you're going to find out what the side effects are.
I think with modern computer systems the effects would be relatively silo'd. I have the sense that each gene are more multi-use and interconnected than execution memory.
The more complex a system is the more reuse of code there generally is. (i.e. multiple different functions making use of some shared function) That means if you randomly alter something, it's more likely to have multiple effects rather than a single effect.
BTW, most times in software parlance "silo'd" means "not interconnected" rather than "interconnected".
