Another look at the Cau et al. 2015 theropod tree – part 1

Most of the time
the large reptile tree covers a wider gamut of taxa (628) with more characters (228) than the smaller more focused studies it is often compared to.

Not so
with a recent study of the Theropoda.

Andrea Cau, Tom Brougham and Darren Naish (2015)
took a recent look at the phylogenetic affinities of theropods using a modified set of 120 fossil theropod taxa and a whopping 1549 theropod characters gleaned from Lee, Cau, Naish and Dyke 2014. That list of characters greatly outnumbers the list in large reptile tree, which nested a mere 53 fossil and extant theropods (including birds) with 228 characters generalized to apply to most reptiles.

Of great concern, 
the tree topologies do not match each other in every detail when pruned to more closely resemble each others’ taxon list (Figs. 1,2). You have to wonder, why do these differences exist? The usual answer: “taxon inclusion/exclusion,” cannot apply in this case. Inaccurate scoring is another possibility.

Cau, Brougham and Naish 2015 reported
“The modified Lee et al. (2014) analysis recovered 1,152 shortest trees of 6,350 steps each (CI = 0.2672, RI = 0.5993). The strict consensus of the shortest trees found is in general agreement with the Maximum Clade Credibility Tree recovered by Lee et al. (2014), the most relevant difference being the unresolved polytomy among Aurornis, Jinfengopteryx, Dromaeosauridae, Troodontidae and Avialae.”

I employed the Cau, Brougham and Naish 2015 data 
and (so far) have included only the first 518 of the complete set of 1549 characters. That’s still more than twice as many characters as in the large reptile tree, and all are specific to theropods. I stopped at 518 because, at that point, all the body parts were listed in order from skull to toes. Thereafter the next 1000+ characters appear to have been tacked on in no particular order. Hopefully I will have an opportunity to add those characters in the next few days or weeks.

With 518 Cau, Brougham and Naish characters
I ran PAUP until it recovered 500+ MPTs and then stopped the analysis. I did not wait for the 1,152 MPTs reported by Cau, Broughan and Naish.

As a quick reminder
the large reptile tree recovered a single, fully resolved MPT and the Theropod subset was similarly fully resolved based on fewer taxa and fewer characters (Fig. 2). More on this below.

Thinking that poorly preserved taxa
might be causing the loss of resolution in the Cau, Brougham and Naish tree, I deleted all of their theropods not duplicated in the large reptile tree (and I made their Syntarsus stand in for my Coelophysis.)

Then I ran another analysis
The most parsimonious consensus tree was still not fully resolved (Fig. 1). Two additional permitted steps produced (also in Fig. 1) recover ever greater loss of resolution – and this with more than twice as many theropod characters, Unfortunately there was lots of lumping in the Cau, Brougham and Naish study, and not much splitting.

Figure 1. GIF animation of the shortest strict consensus tree using 518 characters from Cau, Brougham and Naish and theropods from the large reptile tree. Two other less resolved tree are shown demonstrating the lack of strength at several nodes.

Figure 1. GIF animation of the shortest strict consensus tree using 518 characters from Cau, Brougham and Naish and theropods from the large reptile tree. Two other less resolved tree are shown demonstrating the lack of strength at several nodes.

The theropod subset of the large reptile tree, 
employing only 228 traits, remains fully resolved with a different topology (Fig. 2). The high bootstrap figures tell you that the tree topology remains largely fully resolved even when higher tree lengths are permitted. Smaller numbers appear at weaker nodes. None of the Bootstrap scores are in the 50s or 60s and only a few are in the 70s.

Figure 1. the nesting of Tanycolagreus in the large reptile tree (628 taxa).

Figure 1. the nesting of Tanycolagreus in the large reptile tree (628 taxa).

 

In future blog posts
I will attempt to dig deeper into this quandary, trying to figure out why the tree topologies differ and why there is less resolution in the Cau, Brougham and Naish tree despite the higher and more specific character count. The devil is obviously in the details. I have no idea what I will eventually find here.

Thanks to Andrea Cau
for responding to my requests for data and characters. I will try to treat all aspects of this problem fairly and seriously without animosity.

References
Cau A, Brougham T and Naish D. 2015. The Phylogenetic Affinities of the Bizarre Late Cretaceous Romanian Theropod Balaur bondoc (Dinosauria, Maniraptora): Dromaeosaurid or Flightless Bird? PeerJ. 3: E1032. DOI: dx.doi.org/10.7717/peerj.1032

 

 

 

Advertisements

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out / Change )

Twitter picture

You are commenting using your Twitter account. Log Out / Change )

Facebook photo

You are commenting using your Facebook account. Log Out / Change )

Google+ photo

You are commenting using your Google+ account. Log Out / Change )

Connecting to %s