Completed on 31 May 2018 by Scott Taylor.
Login to endorse this review.
Prost et al. use whole genome sequence data from five bird-of-paradise species to explore patterns of evolution across the birds-of-paradise compared to other passerines, and identify a list of candidate genes that may have contributed to the diverse phenotypes seen within the radiation.
This paper is well-written and clear, but quite speculative and would perhaps be nicer as a shorter report with a reduced discussion. Realistically, an exploration of this nature will always feel speculative and like a bit of a gene list, and the authors acknowledge that the resources they are providing in this paper could set up some very interesting future studies. I agree with this. I think that the methods that the authors have used are sound, and that the genes they have identified are interesting new candidate genes to explore with respect to the incredible radiation that is the birds-of-paradise.
Regarding the discussion of TEs, I agree with the authors that the inclusion of a female in their study likely resulted in this higher repeat density in Lycocorax pyrrhopterus than the other species and I would caution the authors on inferring too much based on this result until they have comparable whole genome data from a male of this species.
Page 2, Line 52 - As written "…more or less strong sexual dimorphism" is a little unclear. You're saying that, in general, these clades generally exhibit strong sexual dimorphism, right? I would reword for clarity.
Page 7 Line 21 -Is there really no way to correct for multiple tests here? Seems to me these could all just be random given the sample size.
Page 9 Line 19 - I would say pigment-based rather than pigmentary
Page 10 Line 48 - 52 - how could positive selection not have played a role in morphological evolution? This seems like a throw away section.
Page 11 Line 20 - 31 - the papers that have not found co-evolution between coloration and vision have examined opsin genes almost exclusively, not the genes found to be under positive selection in this paper. I agree that you're speculating here, but I think the connection between selection on vision genes and the displays and coloration of the birds-of-paradise is probably real.
Page 12 Line 4 - I don't know that the authors should be so literal in their interpretation of the startle response results. Leks are more obvious, yes, and so predator evasion is important (What eats an adult bird of paradise?). Startle response is probably also important for interactions with other males in a lek and with visiting females. I would rather this part of the discussion was not just focused on the higher predation likelihood and more holistically discussed startle response in the context of bird-of-paradise lek behavior.
Page 12 Line 22 - I think the "Other positively selected genes and enriched GO categories" section could be much reduced. Include the function of genes in the list of genes under positive selection rather than including that in the main text here.
Page 17 Line 39 - excess, not access?
Supplementary Table 3 - do intron size and number really not vary between the genomes?