Cellular Mutation

Overview

We tend to think that humans have one consistent, uniform DNA code throughout all cells in the body. One question I have wondered about is, to what degree do all cells of the body actually share the same DNA?

Rate

There’s some muation rate, after all, when transcring DNA. We know that mutation rates are roughly $2.5\times10^{-8}$ mutations per nucleotide site. The robustness mechanisms for transcribing DNA turn out to be quite good, because this rate is actually shockingly small. For some context..

Human Genome Size

The human genome is roughly 3 billion bp, of which only 1%, or 30M bp are part of the exome. On average, we would expect 75 mutations per whole-genome replication, of which (assuming uniform randomness of mutation) just 0.75 of these mutations (on average) would end up expressed.

Cells in Body

Further, by volume, a typical adult human is on the order of $0.062m^3$ - and if the skin cell is representative of an average cell, the average cell is roughly $(10^{-5}m)^3$, making a human contain an order of 10 trillion cells ¹. Given the mutation rate, all cells in the human body would contain just 250,000 mutations. Of these (assuming uniform randomness of mutation), only 2500 mutations would be expressed in any cell’s exome in the entire human body, at a given time; of those mutations, it’s not even guaranteed that the exome would be expressed in that particular cell as part of its transcriptome (since a given cell only runs part of the genetic code).

Biology by the numbers actually verifies the $10^{13}$ number when estimating by volume with similar methodology.²

Pointwise Mutations are Shockingly Low

I think now I finally get the sense for how shockingly low the mutation rate actually is when transcribing DNA. 2500 mutations in the entire body of 10 trillion cells is forgettable.

Caveats

Of course, there are a few simplifications in this argument, that could change the conclusion:

• First, a static snapshot of 10 trillion cells, is, in practice, too low, because cells regenerate. For instance, red blood cells, which are small, make up roughly 84% of human cells, and regenerate roughly 3 times per year - that means that if a human lives 100 years, the exome mutations can accumulate at a rate of 2100 per year from red-blood cells alone.
• Second, and crucially, not all mutations are equal. If a mutation hijacks the replication circuit of a cell, to make the mutated cell reproduce at a rapid rate, the muation can explode in numbers and come to dominate other, well-behaved cells that replicate a normal, smaller rate.³ For example, one explanation for glucotoxicity of $\beta$-cells (which causes diabetes), is that the glucotoxicity actually adds robustness to ensure that overly-sensitive, mutated $\beta$-cells die before they can reproduce and tank blood sugar to deadly lows.
• Third, mutations are probably not uniformly random. For example, enviromental carcinogens can expose certain cells to dramatically more harm than others, and the mutation rate, therefore, likely increases significantly for those cells.
• Fourth, Genome rearrangement can also provide a significant mechanism for evolutionary change to a genome.