Non-coding DNA correlation with rate of evolution

Question

Non-coding DNA can be helpful in generating useful mutations that can go on to become new features/functionality of an organism. Non-coding DNA also indirectly reduces chances of mutation of useful/functional genes.

So if an organism has a larger proportion of non coding DNA it'll be safer from mutation. Humans have 80%-90% non-coding DNA. But this number is far lesser in many other species especially plants. Some plants have as few as only 3% non-coding DNA.

My question is, do these species with less non-coding DNA undergo evolution faster? Because their functional genes are more susceptible to mutations (which would be disadvantageous on average I guess). Would it be fair to say that evolution tries to keep expanding this proportion of buffer DNA? Lastly, is my question related to C-value paradox?

---

My assumption that non-coding DNA indirectly offers adaptive advantage (evolutionarily speaking) seems to be questioned. Let's consider the following case:

Let A and B be two genomes of equal size. Let A have 80 % non-coding DNA and 20% coding DNA. And let B have the opposite i.e., 20% non-coding DNA and 80% coding DNA. Let both A and B be exposed to the same amount of UV radiation for the same amount of time. All conditions being the same, let's assume that this causes the same amount of mutation in both A and B, say 10%. Now what is the probability that this mutation happened in non-coding DNA (which means it doesn't harm the organism) in each of A and B? Evidently, the probability is higher in A meaning it's safer from mutation caused due to UV radiation. So A has an advantage of surviving over B.

Answer 1

Whether or not species with less non-coding DNA evolve faster, the assumption that this could be explained by their genomes being more susceptible to mutation would seem to me to be based on a false premise.

Let us examine the logic of the assumption by making an analogy to drawing of tickets from an urn in a lottery. If you buy one ticket (coding region), and only one ticket is drawn (there is a single mutation) from the urn (the whole genome), the probability of winning (your gene suffering a mutation) decreases as the number of tickets (the size of the genome) increases.

So the question is “does mutation follow this model?”, i.e. are the number of mutations in a genome independent of its size? To answer this, let us look at two examples of known mechanism of mutation.

• Misreading during DNA replication: Here (and in other types of misreading) the error frequency is, quite logically, related to the number of bases replicated i.e. it has a specific (if very low) value of n per 1,000,000 bases — not n per genome.
• UV radiation: Here the source of the radiation is external and its flux will be proportional to the area it strikes — hence the larger the target area the more interactions with thymine causing dimerization. Although our linear-centric view of DNA tends to ignore its area, the larger the genome, the larger the target. The only way I can see the non-coding DNA buffering the coding DNA from radiation is if it somehow formed a protective shell around the latter. I know of no evidence that this is so.