Might it be a bit like unused or commented out functions in computer code? The work to remove doesnāt benefit the system directly so they never get taken out even if itās technically more resource intensive to retain them over time, in the absence of pressure to remove they just stay. Theyāre still functional just not required anymore, but a simple change could drop it back in, even if accidental / unintended.
And unused code does have effects on the system as a whole, itās maintainability, the ease with which you can describe the system etc. but if you just treat code like a discrete thing separate from itās use then maybe not
That would be if it was (mostly) vestigial but itās even simpler than that.
Say thereās a sequence like the first one above, a coding region. Then thereās 1000 C. Then thereās another coding region. If you conceptualise DNA as being a linear line of code, then thatās two unrelated coding sections separated by a load of junk.
But a DNA molecule in your cell isnāt a straight line. Itās a very tightly wound ball of yarn. Itās spaghetti wrapped around a meatball. So, in fact, those two coding regions might actually be adjacent in physical space, even with 1000 base pairs between them, and that might be crucial for them to code what theyāre supposed to. And in that case, the function of the 1000 C units is to provide a flexible backbone linking the two coding sections together in space, that can reversibly attach from the meatball when the coding regions need to be read together.
The 1000 C units are never going to be a coding region. But they have an absolute vital function, based on their physicochemical properties, and theyāre certainly not ājunkā or useless.
Although there is a good coding analogy here - it can be that the second coding region is commented out by default, and the function of the first coding region is to remove the block comments under certain conditions, that the ājunkā DNA sequence responds to.
