Re: Kerberosaurus manakini

[Mickey Mortimer <Mickey_Mortimer111@msn.com>]

Jaime Headden wrote-

>   I think a slight misunderstanding here is occuring. The nature of the
> arrangement of characters may only represent the intuitive neccessity of
> placing characters that diagnose clades one is aware of in a series. I
> certainly have done this for my oviraptorosaurian analysis, in which
> characters are arranged in a hierarchy. I represent possible characters
> diagnosing alternate clades or other possible taxon arrangements below the
> clade that I assume to be more correct by previous analyses. Below all
> this are other characters I treat as being variable among all taxa, or
> general characters. Characters including in the subsets, as in Sereno's
> "sorted" character lists, are supported by the "intuitive" knowledge they
> should be there to help "support" a grouping, even though I add states or
> split characters that I am aware can diagnose other clades, or seek to
> reduce the effective use of that character. Characters 1-24 in my analysis
> are used to diagnose Oviraptorosauria alone. I also separate stem and node
> characters for other reasons. This sorting does not make the analysis
> intuitively predetermining. PAUP* does not care how binary or supra-binary
> characters are arranged, only genetic codings. I understand this can lead
> to assuming a "loaded" analysis, but the data should only have to speak
> for itself. Now imagine the lack of consistent placement for half the taxa
> in one's analysis, and how that is determined. The prevalence of uncodable
> states or high percentage of uncodable taxa below 50% doesn't help,
> either.

I think the misunderstanding is on your part.  I'm not commenting on the
arrangement of characters, which as you state has no affect on PAUP.  If
sorting characters phylogenetically were the problem, there would be no
difference in the Consistancy Indices.  Nor does this have to do with the
number of uncodable taxa.  Sereno's (2000) pachcephalosaur analysis has a
ton of missing data, but is still "loaded" as you put it, with a whole 2
discordant characters out of 574!  Missing data has no affect on CI, which
is an astounding 0.97 for that analysis.

>   As Mickey has shown, however, removing taxa with low percentage of coded
> characters DOES alter the matrix, so there is an effect. However, less
> tested in these analyses is the effect of removing those characters which
> have less than 50% coded positions among all taxa.

No, that was tested by Wiens (2003) as well.  He concluded this generally
has a negative effect on the accuracy.
Wiens, 2003. Incomplete taxa, incomplete characters, and phylogenetic
accuracy: Is there a missing data problem? Journal of Vertebrate
Paleontology 23(2):297-310.
http://life.bio.sunysb.edu/ee/wienslab/wienspdfs/2003/jvp.pdf

> One can get a lower CI
> also by enforcing a topology.

By enforcing a less parsimonious topology, given your data set, yes.

> There are many ways to adversely affect a
> matrix _a priori_ to the set run, but Mickey's complaint seems to be with
> Bolotsky/Godefroit's so-called "loaded" analysis and a predetermination he
> has no awareness of. Sereno's process, unlike that of Godefroit, is
> slightly different, as characters are culled _a posteriori_ (pers. comm.)
> as has occured with most people who run analyses.

Really?  Why would you cull a character a posteriori?  If your matrix finds
a topology despite "bad characters", why erase them from your published
matrix?  It only ends up misrepresenting your analysis.  And why wouldn't
you report this action in your Methods section?

> The particular mindset
> that "every possible character able to be put into the matrix" has another
> adverse problem: inclusion of characters that are considered
> "size-related," while having a likely phylogenetic component (say, within
> tyrannosaurids) does not help the analysis when it could lump the classic
> "Carnosauria" together, despite their exclusion showing tyrannosaurids lie
> closer to birds. Large titanosaurids versus smaller ones? Maybe, maybe
> not. Dromaeosaurids appear to have a trend towards large, terrestrial
> forms evolving from smaller, arboreal/scansorial/aerial forms, so there is
> definately a phylogenetic component to a size increase. Relation of
> humeral and femoral circumference to length of the bone, or position and
> ratio of lengths of moment arms and their bony equivalents, may also
> signal a phylogenetic component, as in hadrosaurids, but these can also be
> size-related. These are culled _a priori_ from most recent analyses and
> are usually discussed as such, but I don't see anyone complaining.

Because there are valid reasons to think such characters are correlated, as
you know.  Sereno is obviously not just culling correlated characters from
his matrix, because these would not affect the CI in most cases.   If two
characters are really correlated, they will both support the same clade.  So
the "extraneous" correlated character will have a similar CI to its partner,
and the resultant CI of the matrix will be similar to a matrix without the
extraneous character.  The only way you could get a higher CI through the
culling of correlated characters is if most of the correlated characters
conveniently were also ones that disagreed with your phylogeny.  Then you'd
be culling characters with low CI's more often than ones with high CI's, so
your total CI (which is an average of each character's CI) would increase.
I can see this happening in some matrices.  I _cannot_ see this happening in
_every single_ matrix Sereno makes, regardless of the number of characters,
taxa or which group he is examining.  So Sereno's culling is not due to
character correlation, which leaves the question of his reason for culling
characters open.

> There is always a bias in selecting characters for inclusion, and
> at least we are given the exact reasons in Bolotsky and Godefroit's
> continuing Kundur/Jaiyin Amurese papers of hadrosaurs. This gives us a
> more capable means of assessing the data that in some phylogenies that
> have been casually posted HERE.

We are given reasons for three such characters, yes.  And that's definitely
a good thing.  Bolotsky and Godefroit's discussion of included characters is
welcome and helpful.  I never said they did everything wrong.
As for your oh-so-ambiguous reference to my coelurosaur analysis posts, you
hit the nail on the head with your use of the term "casually".  In the
publication of my phylogeny, discussion will be included.  Right now though,
my posts just serve as updates to those interested.

Mickey Mortimer
Undergraduate, Earth and Space Sciences
University of Washington
The Theropod Database - http://students.washington.edu/eoraptor/Home.html