I wonder how many simple ordinary mutations (not necessarily codon-affecting errors) are in the final E. coli genome? Any process like this... Probably quite a few, all of which would be sand in the gears.