[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index][Subject Index][Author Index]

Re: Re: Re: Holtz vs. dino dna

Tom Holtz writes:
>These statements, if nothing else, shows that Chip needs to get up to date
>(like, about the 1960s, at least!) on vetebrate paleontology.<<

Yes Tom. you are correct. I am reading a book from 1945 by Romer entitled
Vert. Paleontology.  I have not the time nor the resources(is there a
Medline like computer database for paleontology? If so, let me know.) to
gather in up-to-date articles, other than read Nature and Science weekly.. I
browse the library at OSU for books that have been referenced in articles.  

I hope you enjoyed the SVP meeting, although I picked this up....
>>In any case, Sereno & I disagreed about some character coding, and of course
he's seen MUCH more of the Eoraptor and Herrerasaurus material than me (though
I hope to rectify that when some grant money comes through),<<<

What is character coding in your field?  How many morph character states do
you use in your comparisons for taxonomic purposes?  If you listed the
noticeable differences between Eoraptor and Herrerasuarsus, how many would
there be?

With gene sequencing, for instance, the 18S rRNA gene, there are upwards of
120 phylogenetically informative sites in the 1880 base pair gene.  That's
just that gene. Multiple gene analysis can increase the statistical
likelyhood of the phylogenetical analysis being correct by a lot!  However,
studies have shown that molecular evidence is best evaluated at the single
gene level, putting emphasis on those genes for which the largest number of
taxa are represented.  This is the basis for my outre comments about
morphology.  I can't imagine that you have enough skeletal material from
enough taxa to come up with a hundred or so phylogenetically informative
sites based on morphology.  Correct me if I am wrong!

This is where I may be able to help you guys out, like with the Eoraptor
comparisons.  I can target genes that are species specific, and perhaps help
you resolve identification and classification issues- if warranted!

Now, you replied to this missive....
>Here's an interesting thought.  What would happen if someone was to take a=
> cladogram, and superimpose it onto the geologic time scale (a combination=
> of cladistics and phylogeny)?  Would this be informative, or merely=
> garbage?  At first glance, it could be a powerfull tool.  One could use=
> cladistics to see how animals are related (something that the time scale=
> could obscure), and then put in the timescale, where it could be more meani=

>>The graphic you just described (i.e., a cladogram mapped ontp stratigraphy)
IS a phylogeny.  In someways it is more meaningful, indeed!  Among others,
it allows for putting a putative ancestor directly ancestral to the
hypothesized descendants.

Thomas R. Holtz, Jr.<<

Well, doesn't this sound familiar, exactly what I do! Predict a common
ancestor and date it!  Except instead of using stratigraphy, I use a
molecular clock, which seems to be the problem, some people refuse to
believe such clocks exist.  Here are a few refs you might want to look at:

1. Tetrapod Phylogeny Inferred from 18S and 28S Ribosomal RNA Sequences and
a Review of the Evidence for Amniote Relationships.  Hedges, S.Blair, et.
al. 1990. Mol. Biol. Evol. 7(6):607-633.

2. Inferring Phylogenies from mtDNA Variation: Mitochondrial Gene Trees
Versus Nuclear Gene Trees. Wm. S. Moore  1995. Evolution 49(4):718-726.

3. Phylogenetic Relationships Among Extant Classes of Echinoderms, as
Inferred from Sequences of 18S rDNA, Coincide with Relationships Deduced
from the Fossil Record. Hiroshi Wada and Noriyuki Satoh. 1994. J.Mol.Evol.

4. Details of the Evolutionary History from Invertebrates to Vertebrates, as
Deduced from the Sequences of 18S rDNA. 1994. Hiroshi Wada and Noriyuki
Satoh. Proc. Natl. Acad. Sci. USA 91:1801-1804.

You said....
>>To turn this into a constructive comment, perhaps you can test your very
unconvential phylogenetic hypothesis with DNA fragments from extinct (or
living) pseudosuchians, lepidosauromorphs, anapsids, and nonmammalian
synapsids.  At least this has the potential to compare these different
lineages with mammals, dinos, and birds.  I can direct you to abundant
sources of all of these.  In fact, the results of such a study might well be
worth a presentation at the North American Paleontological Convention
symposium on preserved biomolecules!<<

I would be more than pleased to look at the material you allude to.  You
have a far better understanding of this material, and how it relates to
evolution, so if you could chose which pseudosuchians, lepidosauromorphs,
etc. that would be the most useful, and obtain some, send me some bone
fragments, the size of a peanut will do, and code them so I test them blind,
like everyone else who sends me samples, I will run them through the mill!
I'll sequence the gene products I get(if any, that's a big if!), have the
computer generate the trees, and then let you 'fill in the name' thing.  If
the results contradict what is hypothesized to date, then we need a serious

I have, hitherto, dealt largely with bacterial genetics, especially vector
borne diseases like Borrelia burgdorferi(Lyme Disease) and Rickettsia
(Spotted Fever, et. al.).  When dealing with organisms that do not form
reproductively isolated clades, I have to get to the nitty gritty, genes
that can differentiate species in the absence of observable reproduction and
other characteristics. Not easy!  The neat thing is, most of these genes can
be used in the same office for vertebrates.  Therefor, I have a habit of
ignoring phenotype.  I was part of the group that discovered two additional
species of Borrelia in Europe that also caused Lyme Disease, hence the
nomenclature now reads Borrelia burgdorferi sensu strictu, or sensu lato, or
B. garini and B. afszelli(I hope I spelt those right, don't have the
articles at hand).  These species can only be clearly differentiated, as far
as I know, (I gave up on that field years ago when it became too
political!), genetically.  Perhaps I shouldn't carry this attitude over to
other fields. I apologize if I have ruffled some feathers. Like I alluded
to, I test things blind, isolate the DNA, PCR amplify genes or parts of
genes of interest, sequence them directly(no cloning here!), and computer
analyse the data. Then I label the results. The trouble is, what I have
related is what I got, and that is that among frogs, snakes, salamanders,
iguanas, birds, humans, rats, mice, rabbits, protists, algae, fungi....,
dinos and mammals clade together.  According to a  paper by Blair
Hedges(referenced above), 18S rRNA, myoglobin and B hemoglobin gene sequence
analysis indicated a close relationship between birds and mammals.  This was
using whole gene sequences, and Blair Hedges did not have dinosaur sequence
for that gene..  I simply have not, and maybe will not be able to get entire
dino gene sequences. Also, the informative characters that link mammals and
birds in Hedges analysis are in parts of the gene I have not been able to
recover from the dinos that I have. The parts of the gene that I can analyse
for a theropod and an ornithischian, show a closer relationship to mammals
than anything else for which there is sequence.  In my analyses, I include
only those segments of the gene from extant organisms that are contiguous
with the dinos.  For instance, I analyse only bases 60 through 283 of bird
18S, dino 18S and other extant 18S sequences available. Using only this
region of the gene, the birds diverge earlier than do the mammals and dinos.
This segment of the gene is not exactly representative of the whole gene,
but is good enough to show a trend.  Since, using the whole gene, birds and
mammals form sister clades, and using a partial sequence, dinos and mammals
form sister clades, it may be that all three are linked in some fashion.  

You're probably asking yourself how birds can be related to mammals.  You
are doing the same thing you are accusing me of, assuming a priori that your
morphological comparisons are more important.  What you should consider is
that at a physiological level, birds( and dinos) may indeed be more closely
related to mammals- ie. warm blooded, bipedal posture in some cases,
insulated(hair, feathers, whatever!) than reptiles or fish, and sharing
other similarities that may not be immediately apparent.  Consider the
monotremes, egg laying mammals, with, I believe(again correct me if I am
wrong), nucleated red blood cells.

Well, I'll stop here and anxiously await the replies.


PS- by the way, I'm most curious about Ruben's respiratory turbinates. Have
you any references dealing with those features? What can you tell me about them?