Re: For Sean Pitman: Some real nested hierarchies
- From: John Harshman <jharshman.diespamdie@xxxxxxxxxxx>
- Date: Tue, 11 Mar 2008 08:10:13 -0700
Glenn wrote:
On Mar 10, 11:34 am, John Harshman <jharshman.diespam...@xxxxxxxxxxx>
wrote:
Here is a protein-coding region, tas2r38, which codes for a tasteI'm aware that there is variation among a single species. Here you
receptor protein. The sequence is 1002 bases long, but is highly
conserved; only 40 of them vary among species here, and only 10 are
relevant to a nested hierarchy. All 10 are shown below:
Homo_sapiens cggtatccag
Gorilla_gorilla cggtatccga
Hylobates_klossii tgacaaccaa
Pongo_pygmaeus tgacaactga
Pan_troglodytes cagtgttcgg
Pan_paniscus cagtgtttgg
These 10 sites are enough to uniquely determine a tree, shown below:
seem to be representing each species with a certain sequence,
apparently not only these 10 bases, but all 1002. Is it really true
that all these species have "a" gene of exactly 1002 bases that code
for "a" taste receptor protein?
No. The gene is a bit longer, though what I've analyzed is the entire coding sequence. And yes, it's exactly the same length in all these species, though it might vary in length in other species not studied here.
And why are only 10 bases out of 40
different bases between species relevant to your claim of evidence of
a nested hierarchy?
Because they're the only ones that are informative of relationships. In order to be informative, there must be one base in at least two species and another base in at least two other species. Other patterns tell us nothing. If, for example, the gorilla had an A where all other species had a G, all that tells us is that the gorilla is different from all the others, which I hope we already knew. If, on the other hand, the gorilla shared an A with the orangutan, and everyone else had a G, that would be evidence that the gorilla and the orangutan shared an ancestor to the exclusion of the other species.
Is it possible that a certain percentage of homo sapiens population
have the same sequence as gorilla, cggtatccga?
No.
Don't you need to know
the answer to this?
Yes and no. See below.
And if so, could you direct me to the database so
that I can see for myself one way or the other?
I'm sure there are quite a few human tas sequences in GenBank, and you can search for them. But it's hardly necessary. Human genetic diversity is much lower on average than would be required for any such thing. And anyway, a random sampling of polymorphism would not to produce the same result consistently. Remember that I repeated this analysis with another gene, and that I could have done the same thing with any gene whatsoever, with the same result.
Now there is one situtation in which your idea could be right for single genes. A few genes show strong selection for diversity, such as some of the immune system genes. In those cases, separate alleles are maintained within populations for millions of years, and there are, for example, some chimp alleles that are more closely related to human alleles than to other chimp alleles. Pick a combination of such alleles at random from several apes and you could easily end up with a seriously wrong phylogeny. But like I said, you wouldn't get the same phylogeny twice, much less hundreds of times. So we're on pretty safe ground here.
.
- References:
- For Sean Pitman: Some real nested hierarchies
- From: John Harshman
- Re: For Sean Pitman: Some real nested hierarchies
- From: Glenn
- For Sean Pitman: Some real nested hierarchies
- Prev by Date: Re: programming bomb-testing experiment on a regular computer
- Next by Date: Re: Sean Pitman and nested hierarchy
- Previous by thread: Re: For Sean Pitman: Some real nested hierarchies
- Next by thread: Re: For Sean Pitman: Some real nested hierarchies
- Index(es):
Relevant Pages
|
