RE: molecular clocks: (was Re: Letter from Reasons to Believe)

Hi Wayne,
David can probably address these questions much better than I but I can give you a couple of relevant examples. I've worked with several gene sequences for the purpose of reconstructing land plant evolution. Currently I am engaged only in unravelling the relationships in a smaller group, the hornworts. Each present their own challenges. The latter presents a challenge because I work with chloroplast and mitochondrial genes. In many groups of plants these genes can be efffected by a process known as RNA editing. RNA editing in plants occurs when the transcript of a gene experiences alteration of some Cs to Us and in the case of hornworts Us to Cs as well. As a result the final trancript that is translated into a protein actually has a different code than was present in the original DNA strand. These edited sites are usually important for restoring the function of the protein and so effect 1st and second base positions in the coding strand. In hornworts I work with genes (rbcL and nad5) that may have as many as 8 stop codons in one species all of which are "fixed" by the RNA editing process. Aside from being a fascinating biochemical process that is poorly understood RNA editing can clearly effect the clicking of the apparent molecular clock in these sequences. For example, I have one hornwort taxon that represents a potentially basal hornwort lineage that has very very low rates of RNA editing. As a result its sequences code for fully functional proteins which are very similar to other land plant lineages. Other hornwort taxa may have 20-55 edited sites in a 1000 bp gene sequence. Nearly every one of these edited sites results in the "wrong" base in the DNA code. As a result these samples appear very very different than other land plant lineages. The molecular clock is no good for determining times of divergences because if the mechanism that enables RNA editing has a mutation that makes it more effecient then the sequences can quickly gain mutations because the editing process will fix. Without editing these mutations woudl certainly have been lethal. If the editing mechanism were to have a mutation itself that causes it to become less effecient then there will be high selection for back mutation to get rid of editing sites and restore the original good DNA code. We don't know if this is happening but this is one of the hypotheses that I've proposed or the divergent rates of RNA editng I've seein in my hornworts sequences.

These edited sites effect phylogenetic reconstruction because if two sample have high editing rates but are not related they may share many of the edited sites by virture of the fact that certain position within a gene may be more likely to be edited (again this has yet to be tested). This may cause these two samples to converge in some forms of analysis. If the editing efficiecy changes over time in different land plant lineage then this will cause a change in the observed mutation rate in these genes in these lineages.

This is only a single example of how genes in different lineages may be evolving differently. As an aside I've always wondered by ID proponents haven't lacked onto RNA editing as an example of irreducible complexity". if you think about it, RNA editing is quite a mindboggling process where these plants ahve to "know' that the DNA codes they possess are wrong and must edit them to the correct bases prior to making proteins. The process itself isn't that hard to imagine, although we dont' have a good grasp on it yet, but how the system evolved is a real brain twister. algae don't possses this ability and so the system evolved in land plants. How did the first plant figure out how to reverse a mutation at a specific location and then remember to continue doing this every after. I have some ideas which I can share if anyone is intersted.

Joel Duf
Akron OH

-----Original Message-----
From: asa-owner@lists.calvin.edu on behalf of Dawsonzhu@aol.com
Sent: Thu 5/25/2006 8:12 PM
To: asa@calvin.edu
Subject: molecular clocks: (was Re: Letter from Reasons to Believe)
 
David Campbell wrote:

> I have yet to encounter convincing evidence that molecular clocks can be
> reliable enough to give precision like that, despite the persistant use of such
> calculations by molecular biologists.
>
>

This is quite interesting. Would you be willing to say
a little more on this? Does this extend all the way to
neutral mutation? Are these differences species depended?
What might drive the differences and would the rates
actually make a factor of two difference over millions of
years? Indeed, what error bars should we expect or be
prepared to accept in such analysis?

Perhaps it is also a question on how one measures
species diversification.

It probably would make sense that the rates are not
universally uniform. Your on study in living bivalves
includes species that go back some 4 to 500 million
years.

So much for uncritical usage of GCG and so forth....

it's by Grace we learn,
Wayne
Received on Fri May 26 09:16:57 2006