Posted: Jul 24, 2017 10:05 am
by Rumraket
To pick a really extreme example, there are over one million ALU insertions in the human genome, for example. They're old copies of the gene for 7SL RNA.

A tiny fraction of these are known to be functional as transcription factors for known genes. Another small fraction are implicated in disease because of their effect on trancription some times causes them to mess up normal gene expression (usually because mutations have made them "active" again after having previously deactivated to deleterious mutations). The vast majority are degraded retrotransposons.

I think what you're missing is that we actually have a pretty good estimate of the fraction of mutations that are deleterious, as in cause a reduction in fitness. With this variable in hand (and others such as mutation rate, genome size, population size and average fecundity) it can actually be calculated how big the mutational load will be from deleterious mutations, which in turn will tell us how big a fraction of the genome can be functional.

What I've been trying to get across is that even if we disregard the math, we can see on the actual DNA what kinds of genetic elements our genome consists of, and it is consistent with the results implied by genetic load calculations.