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

Re: Hwang et al. dataset runs (long, responding to Mickey and David both)



Mickey Mortimer wrote:
I got 48195 MPT's when using the heuristic search, with the consensus being
identical to that in Hwang et al. (2002).


Did you use the consensus tree command or use the output given at the end of "bootstrap"? I think I'm beginning to see the problem...


Egh, I see now, there's Heuristic Search.  I'm using that now.


Actually, Archaeopteryx isn't basal within Maniraptora in your consensus
tree.  It's just in a polytomy with Ornitholestes, segnosaurs, troodontids,
dromaeosaurids and oviraptorosaurs.

Yes, that's what I'm trying to say, that's a much better way.

When I ran the analysis with only taxa known in Paul 1988, I got 90 MPT's-
|--+--Albertosaurus
|  `--Tyrannosaurus
`--+--+--Harpymimus
   |  |--Garudimimus
   |  `--+--Gallimimus
   |     `--Struthiomimus
   `--+--Ornitholestes
      `--+--+--+--Segnosaurus
         |  |  `--Erlikosaurus
         |  `--+--Avimimus
         |     `--+--Microvenator
         |        |--Oviraptor
         |        |--Conchoraptor
         |        |--Ingenia
         |        `--+--Citipati
         |           `--Chirostenotes
         `--+--Archaeopteryx
            `--+--+--Saurornithoides mongoliensis
               |  |--Saurornithoides junior
               |  `--Troodon
               `--+--Dromaeosaurus
                  |--Adasaurus
                  |--Deinonychus
                  |--Saurornitholestes
                  `--Velociraptor
Looks identical to Hwang et al.'s tree with all taxa included, except for
the Citipati + Chirostenotes clade.


Why did you include _Citipati_, or are you referring to GI 100/42?

I re-ran the analysis. The proper multistate characters are ordered, I did the following.

1. Opened PAUP
2. Opened the nexus file
3. Used the command "HSearch"
4. Used the command "ShowTrees"

Tree number 1 from that run

--+--Allosaurus fragilis
     `--+--+--Tyrannosaurus rex
         |   `--Albertosaurus libratus
         `--+--+--Garudimimus brevipes
             |   `--+--Harpymimus okladnoviki
             |       `--+--Gallimimus bullatus
             |           `--Struthiomimus altus
             `--+--Ornitholestes hermanni
                 `--+--+--Archaeopteryx lithographica
                     |   `-+--Dromaeosaurus albertensis
                     |      `--+--Adasaurus mongoliensis
                     |          `--+--Saurornitholestes langstoni
                     |              `--+--Deinonychus antirrhopus
                     |                  `--Velociraptor mongoliensis
                     `--+--+--Troodon
                         |   `--+--Saurornithoides mongoliensis
                         |       `--Saurornithoides junior
                         `--+--+--Segnosaurus galbinensis
                             |   `--Erlikosaurus andrewsi
                             `--+--+--Chirostenotes pergracilis
                                 |   `--Avimimus portentosus
                                 `--+--Microvenator celer
                                     `--+--GI 100/42
                                         `--+--Conchoraptor gracilis
                                             `--+--Ingenia yanshini
                                                 `--Oviraptor philoceratops

Pretty standard, I suppose. I wonder what would happen in these analyses if more basal birds were included. Has anyone ever tried including _Shenzhouraptor_, _Sapeornis_, etc. in these analyses?


David Marjanovic wrote:
> >Maybe you didn't make the ordered characters ordered?
>
> That would seem not to be the case.  I've checked over the matrix.

That isn't part of the matrix. It's a line in the assumptions block in a
Nexus file.


When you open the Nexus file in Nexus Data Editor, you click "Data" in the menu bar, then click "Character Properties", I went through the list, the characters that need to be ordered are.


> >I'd rather say that there are not enough of these similarities in the
> >matrix -- assuming it isn't a problem of ordered vs. unordered --, and
> >therefore birds and dromaeosaurids can't cluster. And then *D.* lacks
> >something that the other dromaeosaurids have. What could that be?
>
> We don't necessarily need to figure out what it lacks that all other
> dromaeosaurids have, but instead what it lacks that _Deinonychus_,
> _Adasaurus_, _Saurornitholestes_, and _Velociraptor_ have.

That's what I mean by "the other dromaeosaurids". :-)


Mi malo. I thought you meant "the other dromaeosaurids" as in all other dromaeosaurids. Apparently, these dromaeosaurids are united to the exclusion of _Dromaeosaurus_ by the following characters:


13(2). Depression for pneumatic recess on prootic present as a deep, posterolatearlly directed concavity.
187(1). Exits of C.N. X-XII located together in a bowl-like basisphenoid depression.
188(1). Basipterygoid recesses on dorsolateral surfaces of basipterygoid processes present.


Out of curiousity, how can this be known? _Adasaurus_ has braincase material? There's a new one on me.

> > > ...the change in the tree is surprising, I was clearly wrong about
> > > _Confuciusornis_.  Instead, what do we find?
> >
> >And why? :-o

I mean, why does that clade come out, what synapomorphies does *C.* share
with the dromies?


7(1). Subotic recess absent.
74(1). Axial epipophyses large and posteriorly directed, extending beyond postzygapophyses.
76(0). Epipophyses of cervical vertebrae placed distally on postzygapophyses, above postzygapophyseal facets.
77(0). Anterior cervical centra level with or shorter than posterior extent of neural arch.
84(1). Parapophyses of posterior trunk vertebrae distinctly projected on pedicels.
86(1). All presacrals pneumatic.
99(1). Prezygapophyses of distal caudal vertebrae extremely long.
109(1). Acromion margin of scapula anterior edge laterally everted.
113(1). Scapulocoracoid form a L-shape in lateral view, coracoid inflected medially.
119(1). Proximal surface of ulna divided into two distinct fossae separated by a median ridge.
124(1). Metacarpal I less than half the length of metacarpal II.
133(1). Antitrochanter prominent.
142(1). Ischium 2/3 or less of pubis length.
145(2). Pubis moderately posteriorly oriented.
149(1). Femoral head with circular fovea captialis in center of medial surface of head.
165(1). Distal tarsals form metatarsal cap with intercondylar prominence that fuses to metatarsals early in postnatal ontogeny.
168(1). Distal end of metatarsal III with developed gingylmus.
173(1). Anterior emargination of supratemporal fossa strongly sinusoidal and reacing onto postorbital process.
178(1). Frontal edge notched in region of lacrimal suture.
182(1). Internal mandibular fenestra large and rounded.
196(1). Anterior end of ilium strongly curved.
205(1). Shaft of metatarsal IV mediolaterally widened and flat in cross-section.
207(1). Tuber along dorsal edge of ilium, dorsal or slightly posterior to acetabulum.


Re-running the final one I did with Heuristic search, the one where _Confuciusornis_ is included but segnosaurians are excluded produces the following-

--+--Allosaurus
    `--+--+--Tyrannosaurus
        |   `--Albertosaurus
        `--+--+--Garudimimus
            |  |--Harpymimus
            |  `--+--Struthiomimus
            |      `--Gallimimus
            `--+--Ornitholestes
                `--+--+--Avimimus
                    |   `--+--Microvenator
                    |       |--Ingenia
                    |       |--Conchoraptor
                    |       |--Oviraptor
                    |       `--+--GI 100/42
                    |           `--Chirostenotes
                    `--+--+--Archaeopteryx
                        |   `--Confuciusornis
                        `--+--+--Troodon
                            |   |--Saurornithoides mongoliensis
                            |   `--S. junior
                            `--+--Dromaeosaurus
                                |--Deinonychus
                                |--Adasaurus
                                |--Saurornitholestes
                                `--Velociraptor

Excluding _Confuciusornis_ produces the same results, except for that _Archaeopteryx_ is lonely as the sistergroup to the Deinonychosauria.

GI 100/42 and _Chirostenotes_ are sistergroups on the basis on the following:

92(2).  Pleurocoels present on all sacrals.
96(0).  Anterior caudal centra tall, oval in cross section.
114(1). Scapulocoracoid glenoid fossa faces laterally.


Nick Gardner

_________________________________________________________________
Help STOP SPAM with the new MSN 8 and get 2 months FREE* http://join.msn.com/?page=features/junkmail