In article <78vf24$qmh at net.bio.net>,
Sam King <mjkg at dna.bio.warwick.ac.uk> wrote:
>two sialidase genes from streptococcus pneumoniae which are more
>conserved at the nucleotide level (60%) than the amino acid level
>(53%). I thought this was usually the other way round.
>Ive checked the synonymous/nonsynonymous values using jukes-cantor
> Does anyone have any ideas? or other examples like this?
A couple of ideas:
(1) If the nucleotide frequencies are far from 0.25/0.25/0.25/0.25,
as they are in thermophilic bacteria for example, you might see
false evidence for conservation at the nucleotide level because
mutations keep going back to the same favored base composition.
(2) The gene sequence might also have some function as DNA or
RNA (binding site, etc) which keeps its nucleotide sequence
(3) There could be diversifying selection on the protein, so
that mutations which change the protein are preferred over
ones that don't. (But your synon/nonsynon values don't particularly
support this idea, if I'm reading them correctly.)
(4) The difference could be due to chance, especially if your
sample is small.
Mary Kuhner mkkuhner at genetics.washington.edu