A Phylogenomic Supertree of Birds

Round 1

Reviewer 1 Report

This is an interesting paper. I am satisfied with this manuscript in its current presentation and I just have minor comments/suggestions.

 

You mentioned that the criterion used are “quite fast given the sizes of the data matrices used in this study”, including the ML criterion for the MRL approach. However, I think that there are other (fast) statistical methods (such as ASTRAL -which uses a coalescent model-) that can be used for comparison. I am not suggesting that you need to incorporate these analyses but an explanation about why you chose “only” MRP and MRL and not others (such as ASTRAL) would be helpful. This is important especially to understand biological processes such as ILS that could be detected in early neoavian diversification of the desired supertree.

 

 

Figure 6 is wrongly labeled. The caption said “Figure 1”.

 

One might think that the estimation of branch lengths and time tree will be pushed forward by using mitochondrial genes. Some words that condition (or approaches used to minimise this) on this point would be helpful.

Author Response

See attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

I must preface my comments by saying that I took this paper on mostly out of interest in the findings - I am not a specialist in phylogenetic analysis, so please interpret my (limited) comments accordingly. Overall I found it a good read, and well written considering the rather technical nature of the contents.

The weighting of different source trees (lines 241-253) appears to be rather arbitrary and it would be interesting to explore how changing these weightings alters the resultant trees. The authors acknowledge this in the Discussion, but don’t do anything to assess the impact of the weightings chosen. Varying the weightings needn’t be an exhaustive process, but it would be interesting to explore some of the more controversial groupings/nodes in this regard. I’m concerned because the extra weighting given to densely-sampled genera might have undue influence on the resultant larger scale structures that are of greatest interest. An alternate approach might be to subsample the taxa from very well represented genera (e.g. place an upper limit of ~5 species per genus)

Lines 603-606 – I think that someone’s suggestions for rounding out the Discussion haven’t been acted on!

 

Minor points

Line 41 – megaphylogies should be megaphylogenies

Line 196 – d)ata should be data

Fig. 3 has Accipitriformes as owls, not raptors/hawks

And pelecans should be pelicans

Most people recognize 2 species of ostriches, so give the common name as a plural (as done in Fig. 5c)

‘landfowl’ is a weird term – usually refer to Galliformes as gamebirds?

Charadriiformes are better labeled shorebirds, gulls & allies?

Fig. 5 – use plurals for Painted snipes, Seedsnipes as both have multiple spp.

Standardise use of capitals for common group names (lower case in Fig. 3 but capitalized in Fig. 5)

Line 408 – insert space after (tinamous), and use buttonquails not hemipodes if following IOC nomenclature (also line 525)

Line 448 – insert ‘to’ in ‘was not sufficient to produce…’

Lines 490-493 – something has gone wrong here – the sentence doesn’t make sense

Line 519 – ref should be [43], not Burleigh et al 2015

I suggest including the Appendix and its many associated references in supplementary material

 

I didn’t check the references at all carefully, but spotted a few errors:

Line 1403 two full stops in a row

Line 1461 not tThreskiornithidae (lose the first t)

Line 1489 france should be France

Line 1490 two full stops in a row

Author Response

See attachment.

Author Response File: Author Response.pdf