Genetic Diversity and Population Structure of Two Endangered Neotropical Parrots Inform In Situ and Ex Situ Conservation Strategies

Round 1

Reviewer 1 Report

Summary

The study presented by Campos and co-authors provides the comprehensive description of population structure and genetic diversity of two endangered parrot species: the blue throated macaw (Ara glaucogularis) and the thick-billed parrot (Rhynchopsitta pachyrhyncha). An extensive set of diverse population genetic methods is used to access fine structure of both captive and natural populations. Quantification of genetic differentiation obtained within the present study helps to suggest strategies for conservation and genetic management of both species.

Broad comments

Campos and coauthors present a novel and high quality study. The main novelty aspect is the first access to genomic structure of two populations, which is based on unique data set from native and captive populations. In addition, the present study can have a potential long term impact: the current state of the population described in the present work could be used a reference point for assessing the efficiency of the conservation measures applied in the future.

All presented analysis agrees with the standards of the field. Authors developed and used significant set genetic (microsatellite) markers for both species and thoroughly analyzed them. Inclusion of extensive checks and statistical corrections provides additional confidence in the obtained results and conclusions. Methods, software, and procedures are described clearly and with sufficient details. Overall the writing style and language are fully appropriate.

Quality of presentation can be further improved with addition of connecting sentences throughout the text (see more detailed suggestions in the Specific comments section) and giving broader overview of impact of the results. Additionally, this will help to further improve the accessibility of the paper for readers. In particular, I encourage authors to explain the importance of population genetic approach in conservation studies. Specifically, it would be great if authors provide a more intuitive explanation of what exactly methods and statistics use are telling about overall fitness of the population. Also a comment on the applicability of the results of this study to other species, particularly parrots could help to increase the impact of the present work.

 

Specific comments 

Abstract.

1. I suggest adding a connecting phase after the first sentence stating, that understanding patterns of genetic variation, explaining what do those patterns tell us and preparing reader to further statements about levels of diversity found. How do patterns of divvertiation are connected to conservation status and overall well-being of the population? What is the danger of high level of inbreeding and low diversity in the population? Those aspects are partially discussed in the opening paragraph of the introduction, however, it would be essential to present them even earlier. Also line 90 have some level of perspective there and can be reemphasized in discussion section.
2. Closing sentence: To give more global interpretation: can we say that conclusions can be applied globally to conservation of parrots?

Introduction.

1. I suggest extending this chapter to briefly outline conservation status within parrots: how many of them have conservation status of endangered species? How many species have limited and scattered ranges? This in my opinion could provide a stronger motivation for the choice of the study species, as they seem to represent a typical conservation concerns for entire group of species. It would be beneficial to discuss effective population size in addition to census size.
2. Line 50: Authors discussed one of the possible genetic consequiences of breeding in captivity: increasing levels of genetic diversity. However, there are factors which may have an opposite effect in the captive population, such as founder effect and higher inbreeding levels in uncontrolled cases. I would suggest additionally discussing these aspects.
3. Line 66: What is a “studbook” and how breeding program can influence the study results?
4. Line 95: “Nonetheless, …”

Not clear what is meant by “genetics” here. Do authors imply limited genetic data and absence of full genomes in this sentence?

Methods:

6. Line 118: The dataset was collected during an extensive time period (which is not unexpected knowing difficulty of field sampling). In my opinion a comment on the possibility that population structure may have changed through this time period could be appropriate.
7. Line 185: It is customary to access population structure via PCA plots. What is the rationale for choosing PCoA for this study?
8. Table 1: I would suggest improving readability of the table by placing significance values in the same cell as F_st values themselves (using asterisks for example). Why some F_st values are not available?
9. Figure 2. I suggest omitting this plot. As the text implies, division to two subpopulations is not significant (optimal K=1). However plot may be interpreted as both populations (and both species) have exactly the same number of loci assigned to eighter one or another population, suggesting two distinct genetic pools.
   Related to this plot: is there any explanation why there is no stochasticity?

Discussion:

10. Line 22: the phrase “face many problems” seems to be a bit colloquial
11. Line 47: Because of the citation at the beginning of the sentence this sentence is confusing. Please consider rephrasing.
12. Genetic diversity 4.2, first paragraph. I suggest to start this paragraph with explaining how genetic diversity, population divergence levels and consequences of bottleneck can inform us about population “fitness” and tolerance to conservation threats. What are pros and cons of using population genetic methods for that?
13. Lines 59-65: Comparison between absolute values of heterozygosity is provided. Is difference in heterozygosity estimates (for example for kakapo) expected given the census size?

Author Response

see attached

Author Response File: Author Response.pdf

Reviewer 2 Report

I have made several comments directly onto the manuscript´s PDF. I regard them as relevant issues to be addressed, including these two main ones, highlight below:

1) Make sure you report exactly which index(es) you used to interpret the Structure results. Either within Structure "harverster" or somewhere I cannot remember exactly, you will also fiund the so called "Evanno" statistics, which offers a counterpoint to the ΔK metrics obtained by Structure. Ideally, you should report these two tests for the best value of K, and comment to what extent they converge or diverge. This can make your inferences about the true K among the wild populations more robust in my view.

2) I detected problems with the voucher information on the samples used in the analyses, both concerning captive and wild birds. In the supplementary materials, concerning Table S1: Is there any internal identification numbers for these samples that would allow for the unequivocal identification of each individual. I am looking here for good practices in vouchering specimens used in the genetic analyses. Ideally, these numbers should also show up (i.e., correspond to) individuals shown in Fig. S1. Finally, where is the voucher information on the wild specimens? A separate table should be provided for them, following the same format suggested for Table S1 above.

 

Comments for author File: Comments.pdf

Author Response

see attached

Author Response File: Author Response.pdf

Reviewer 3 Report

See attached Documents

Comments for author File: Comments.zip

Author Response

see attached

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Overall, I am pleased that the authors addressed most comments I made. However, I am still finding the Structure analysis not particularly clear. First, if they tested values between up to k = 10, I see no point in reporting in Fig. 2 that they tested only for K = 2. Also, the plot is still misleading. If no strong evidence was found for K > 1, then there is no point in showing any Structure plots after all, but just to report the test statistics. Finally,  this explanation provided in the review must be somehow incorporated into the results´s part of manuscript: "For both species, the optimal K inferred by STRUCTURE HARVESTER was at K=6 populations as shown by the plotting of ΔK vs K. However, the value of these ΔK were inherently low at 2.2 and 0.6 for the blue-throated macaw and thick-billed parrot, respectively". The most important thing here is to provide a clear rational in the form of a test statistic (maybe?) supporting your decision for disregarding K = 6, which was the population structure value chosen as the most likely by HARVESTER (Fig. S2). Just stating the delta K values were low for K = 6 is not enough, as terms such as "low" or "high" are vague in this case.

Author Response

see attached

Author Response File: Author Response.docx