Genetic and Population Structure of Croatian Local Donkey Breeds

Round 1

Reviewer 1 Report

The aim of the manuscript is clearly fit to the journal. Information about genetic and population structure of genetic resource local breeds could give interesting information.

The introduction part and objectives are clearly described, the chosen methods are suitable to answer aims of the study. Due to the low quality of the pedigree (short generations and too much unknown parents), I suggest to remove those parts of the manuscript. Unfortunately, the short pedigree not allows to compute reliable inbreeding coefficients. Sample size for microsatellite analysis seems to be acceptable, but I feel the sample size (15 animals per breed) for mtDNA analysis too weak.

Question1: How can be a living animal with unknown parent(s) accepted to the Littoral-Dinaric breed?

Author Response

Dear Reviewer,

We appreciate the time and effort you have dedicated providing valuable feedback on the manuscript, and for all the insightful comments. We made changes and the required adjustments in the manuscript.

The introduction part and objectives are clearly described, the chosen methods are suitable to answer aims of the study. Due to the low quality of the pedigree (short generations and too much unknown parents), I suggest to remove those parts of the manuscript. Unfortunately, the short pedigree not allows to compute reliable inbreeding coefficients. Sample size for microsatellite analysis seems to be acceptable, but I feel the sample size (15 animals per breed) for mtDNA analysis too weak.

Dear reviewer, thank you for the suggestion. We are aware that any implementation of the conservation of a local breed begins with the inventory of the local population, the determination of breed standards, and finally the establishment of registries. A similar situation has occurred with donkey breeds in Croatia. Unlike (especially conventional) horse or cattle breeds, the selection of donkey breeds (including the establishment and maintenance of registers) was not the focus of attention in past centuries. Donkeys were the animals of poor people and "archaic" agriculture. Therefore, the pedigree depth of the donkey populations studied is modest, which can also be observed in other relevant and cited studies of donkey breeds (equivalent generations; Amiata donkeys, 0.78, Cecchi et al. 2006; Asinina de Miranda donkeys, 0.28 Quaresma et al. 2014; Andalusian donkeys, 0.75, Navas et al. 2017). Being aware of the weaknesses of some methods (pedigree analysis), we tried to strengthen them by combining several methods (pedigree analysis, genetic analysis of microsatellites and mtDNA; each method has some weakness, considering both contemporarily increase the soundness of the analyses). We believe that pedigree analysis is extremely important in the regular management of small and endangered breeds, and we certainly justify its retention in research as one of the research methods. Through careful selection of females (age, breeding and sampling area, consideration of oral/informal pedigree information), we have attempted to select a representative sample for research on the genetic structure of breeds. We agree with the statement that conclusions are more certain when based on genetic analysis of multiple individuals (multiple mtDNA sequences), but respecting the given constraints (space, resources, population size, etc.) and careful selection of animals for sampling, we believe we have reached reasonable conclusions.

Question1: How can be a living animal with unknown parent(s) accepted to the Littoral-Dinaric breed?

Unknown parents mean unregistered parents (parents can be known at the moment of registration of the animal) and this is true for the first generation enrolled in every Herd Book and also for new animals enrolled later in open Herd Books. In all this cases the decision is based on the phenotype of animals, and sometimes also of their unregistered parents. During and after the inventory of the populations in the field and based on all the information collected (phenotypic characteristics, informal information on the origin, geographical location of the animals, etc.), registers were established and the animals were entered in "open" registers according to established criteria.

We hope that we have succeeded in meeting your requirements.

Yours sincerely,

The authors of the manuscript

Author Response File: Author Response.pdf

Reviewer 2 Report

The aim of this study was to analyze the demographic and genetic status of two donkey breeds after the start of protection by analyzing their pedigrees and genetic structure.

This is a very good and well-written review manuscript. All sections are well described, and discussion is strongly supported. I only suggest considering next minor comments to improve the manuscript:

• Line 97: I suggest to consider mtDNA analysis as a second objective.
• Line 132: I suggest to clarify the procedures used for DNA extraction and genotyping.
• Line 610: Please correct Journal name.
• Line 612: Include the page range.
• Lines 624-625: Only the first word should be capitalized. 
• Lines 660-661: Only the first word should be capitalized. 

Author Response

Dear Reviewer,

We appreciate the time and effort you have dedicated providing valuable feedback on the manuscript, and for all the insightful comments. We made changes and the required adjustments in the manuscript.

The aim of this study was to analyze the demographic and genetic status of two donkey breeds after the start of protection by analyzing their pedigrees and genetic structure.

This is a very good and well-written review manuscript. All sections are well described, and discussion is strongly supported. I only suggest considering next minor comments to improve the manuscript:

• Line 97: I suggest to consider mtDNA analysis as a second objective.

Answer: Thank you for suggestion. Done as suggested, (Line 108-109)

• Line 132: I suggest to clarify the procedures used for DNA extraction and genotyping.

Answer: Procedure of DNA extraction and genotyping are usual in reference laboratories. They are not completely known to us.

• Line 610: Please correct Journal name.

Answer: Thanks to the reviewer for a useful suggestion. Done as suggested, (Line 721)

• Line 612: Include the page range.

Answer: Reference Blasi et al. 2005. is abstract presented on 16^th Nat. Congr. ASPA, Torino, Italy and published in Italian Journal of Animal Science 4 (Suppl.2) on page 127. (Line 722-723)

• Lines 624-625: Only the first word should be capitalized. 

Answer: Thanks to the reviewer for a useful suggestion. Done as suggested, (Line 735-736)

• Lines 660-661: Only the first word should be capitalized. 

Answer: Thanks to the reviewer for suggestion. Done as suggested, (Line 787-788)

We hope that we have succeeded in meeting your requirements.

Yours sincerely,

The authors of the manuscript

Author Response File: Author Response.pdf

Reviewer 3 Report

In this study, Ivankovic and colleagues present pedigree ang genetic data from two Croatian donkey breeds, the Littoral-Dinaric and Istrian donkey breads. They compare both local breeds based on a thorough pedigree analysis. They also investigate 13 microsatellite loci for 60 donkeys of each breed, as well as the mitochondrial D-loop region for a total of 30 individuals. The authors conclude that there is no evidence for bottleneck in these breeds in the last decades, and that they both show high genetic diversity and fall within the diversity of modern European donkeys. Although most of the analyses presented in this study seem sound to me, I have some major concerns that I would like the authors to address.

First of all, the authors almost never explain the parameters they are calculating and using to draw conclusions. I am familiar with most of them -  but not all of them – and I expect many readers to be very uncomfortable with the absence of explanations, especially in the Methods section. I therefore advise the authors to dramatically expand this section, and define in mathematical terms, the parameters they are estimating from the pedigree data.

Second, the authors do not acknowledge the limitations of working with pedigree data. In particular, they estimate many genetic parameters (e.g. inbreeding coefficients, bottleneck) but these can only be estimated within the last ~50 years or so. Importantly, from what I understand, there is no prior knowledge on the relatedness between founders of these two populations. It could well be that they were all closely related and potentially inbred, but there is no mention of this in the manuscript. I would like to remind the authors that an absence of bottleneck in the last decades does not prove that these populations have not undergone some previous bottlenecks in the past. In this regard, such pedigree data cannot give much information on the overall genetic diversity of the populations. I would also advise the authors to restructure the discussion section, as many elements seem redundant. I believe that making the discussion more concise would also make the whole manuscript more impactful.
Third, I find the use of genetic data in this study a little bit misleading. Indeed, there is little comparison of genetic diversity uncovered at the 13 microsatellites in other equine populations, and I do not understand the genetic assignment analysis they present here. Furthermore, there is no mention of any of the caveats inherent to mtDNA analysis (see points below). I strongly encourage the authors to mention these, and expand the mtDNA discussion section, as this represents a powerful tool, somehow under investigated here. They do not comment much on the general pattern of mtDNA diversity observed in these two breeds, nor on how these breeds compare to the rest of donkey populations worldwide.

Fourth, the legend of tables and figures are too short and lack most of the essential information they should contain, which makes it difficult to read and interpret them. I strongly recommend the authors to expand these legends, and edit some of the figures, for the reader’s convenience. In particular, figures 5 and 6 are very confusing, and do not convey any convincing message at the moment.

Finally, I recommend the authors to proof read their manuscript more carefully, as there are still many grammar and syntax issues throughout the main text. I also encourage them to have their manuscript proof read by a native English speaker.

Minor points

35. 16-35. The abstract is too long and therefore not very impactful. In particular, I find it very difficult to identify the key findings and conclusions of the study in this abstract. I encourage the authors to shorten it and only mention the most impactful results.

41. 41. What species are included in ‘mammalian breeds’? Please be explicit. Why mention the number of breeds rather than number of individuals and/or populations?

44. 44. ‘2.2%’: please mention that it is a percentage relative to the number of donkey breeds.

65. 64-65. ‘there were only a few hundred donkeys remained’: this is grammatically incorrect. Please rephrase.

83. 75, 83. ‘others’: which others are you referring to? Please cite relevant literature accordingly.

79. 79. ‘the breed Studbooks was established’: this is again grammatically incorrect.

l.87-93. I am very surprised that the authors do not mention one inherent issue of using mtDNA to reconstruct genetic relationships: it is only transmitted from mother to offspring, and therefore only recovers maternal lineages. Importantly, mtDNA might be very discordant with nuclear DNA, and can provide a biased perspective of the evolutionary history of the species. For instance, in horses, mtDNA analyses indicate a clear lack of phylogeographic signal, and cannot be used for reconstructing genetic relationships between populations.

103. 102-103. ‘some of which were born in 1972’: what about the rest of the founders? When were they born?
     
     
104. 115-121. Please explain each of the parameters that have been estimated: what do they refer to? How are they calculated? Please be very explicit here, for the reader’s convenience.

130. 129-130. Please mention how many males and females of each breed were sampled, and indicate the age distribution of the sampled animals.

141. 139-141. Again, please be explicit here and explain each one of the parameters that have been estimated. What do they refer to specifically? How are they calculated?
     
     
142. 155-156. Again, please mention how many males and females of each breed were sampled, and indicate the age distribution of the sampled animals.
143. 
     196-197. How do you explain such a discrepancy between numbers in REF 01 and REF 02? Why are there so many active animals for which we do not know both parents? I would encourage the authors to briefly comment on this.

203. 202-203. ‘indicating the absence of a bottleneck in the populations’: this only indicates an absence of bottleneck in the last ~50 years. It is impossible based on this data to conclude with regards to a bottleneck prior to the 1970s.

222. 222. Please be consistent in the tense used.

238. 238. There is a word (probably ‘high’) missing between ‘Istrian donkey had a’ and ‘completeness value’.

l.250-251. Again, mind the tense. The authors sometimes use the present tense, and sometimes the past tense, I suggest they stick to one of these two options for consistency.

275. 275. Typo: ‘be-low’ should be written ‘below’.

284. 284. ‘significant excess of heterozygotes of 4.5’: I might have missed it, but I cannot see this value anywhere in Table 3. Can you please mention which parameter in Table 3 you are referring to here?

314. 313-314. I would be very interested to see this FCA plot, as it may provide some relevant insights. Can the authors include it in the manuscript?

319. 319. Why is there a range (91% to 97.2%) in the assignment of individuals in Littoral-Dinaric populations? Please be more explicit and explain the procedure behind this assignment.

323. 323. ‘SMM’. The authors never mention what these three letters stand for, please explain it here.

333. 333. I might have missed it, but how do the authors calculate the mean group distance within haplotypes?

337. 337. Typo: please replace ‘the other country’ by ‘other countries’. Which countries are these other donkeys from?

363. 362-363. Please replace these arrows by words (i.e. ‘from 129 to 158 individuals’).

364. 364. ‘up to ten years old’. What do the authors mean here? Less than ten years old?

374. 374. ‘the female/male ratio was similar’: which ratio are you referring to here? The one in Littoral-Dinaric donkeys or in Istrian donkeys? These two ratios are very different from each other! Besides, how do you explain such a different sex ratio between the two populations?

379. 378-379. I might have missed it, but what is the difference between maximum and complete generations? Please be more explicit here.

386. 379-386. Instead of listing some values for many different donkey populations, why do compare them to the two breeds investigated here? I would advise the authors to rewrite this section, highlighting the similarities and differences.

392. 392. ‘was lower in related on’: this is grammatically incorrect, please rephrase.

407. 497-407. How do the authors explain the fact that there seems to be an age difference for age of parents at the birth of their offspring in some other donkey populations, but not in the two they investigated?

416. 416. ‘in the Amiata donkey 114.9 and 135’: I might have missed something, but how can the effective number of founders be lower than the number of ancestors?

421. 420-421. ‘a larger ratio reflects a stronger bottleneck effect’: please bear in mind that this only indicates a recent stronger bottleneck, postdating the establishment of studbooks/pedigree. Such a ratio cannot give any demographic information prior to the first generation included in studbooks.

427. 426-427. There is a syntax issue with this sentence, please rephrase it.

462. 460-462. I am surprised that the authors do not comment on these 13 microsatellite loci used here. I would advise them to investigate them further, and have a closer look at literature leveraging these loci. What diversity have they uncovered/failed to uncover?

468. 467-468. ‘relatively high degree of heterozygosity in the Croatian donkey populations’: relatively high compared to what specifically? Please cite relevant literature here. In particular, does this heterozygosity seem high compared to other domestic animal species (e.g. horses, cattle)?

492. 492. Typo: ‘lover’ should be replaced by ‘lower’.

499. 496-499. These few sentences are redundant with an earlier section of the discussion. In particular, the authors mention a second time the observed and expected heterozygosity in these two breeds. I recommend the authors to restructure the discussion to make it more concise and impactful.

531. 530-531. Please cite appropriate literature or sources when stating that the ‘population size’ (…) ‘have declined significantly throughout history’. When did that decline happen exactly?
     Besides, based on the data presented here, it is impossible to know whether these two populations experienced a bottleneck prior to the 1970s!

531. 531. ‘these population has’: this is grammatically incorrect, please rephrase.

541. 534-541. I am very surprised that in this section discussing mtDNA, there is no mention of any of the caveats of mtDNA based inferences. I strongly recommend the authors to mention these here, especially as in other equid species, mitochondrial DNA mostly shows an absence of phylogeographic signal.

Figure 1

178. 178. Typo: ‘males, females a new registered animals’ should be ‘males, females and newly registered animals’.

I would suggest the authors to show the trends of newly registered animals for females and males separately, to identify any potential sex bias in number of new individuals identified each year.

Figure 2

192. 192. Typo: ‘population’ should be plural, ‘populations’.

Similar to Fig. 1, it would be great to have males and females separately to identify whether the distributions of both sexes are similar or not.

Figure 3.

227. 227. Add ‘populations’ after ‘Istrian donkey’.

Figure 4.
l. 236. I would suggest to replace ‘populations’ by ‘panels’ or ‘datasets’, and add ‘populations’ or ‘breeds’ after ‘Istrian donkey’.

As the x axis is showing generations, as is therefore discrete data, I strongly recommend the authors change the type of graph here to one appropriate for discrete data, such as bar plot or scatter plot.

Figure 5.

320. 320. Typo: ‘od’ should be replaced by ‘of’.

The legend of this figure does not mention which visualization method is shown in the figure What do the x-axis and y-axis scales represent? Does each dot represent one individual? Please be more explicit in the legend.

I would be also very curious to see on this figure the few individuals that have not been assigned to the population they were sampled from.

Figure 6.
The legend does not mention which type of phylogenetic inference is shown. Is this tree based on maximum-likelihood, Bayesian, distance or maximum parsimony? What does the scale, colours and shapes on the tree represent? Why is there no support for the nodes? Please be more explicit, as it is difficult to interpret placement of the two Croatian breeds without additional information.
Besides, I would recommend the authors to represent a network of haplotypes, with number of substitutions separating haplogroups, to identify the geographic diversity represented in each haplogroup. 

Author Response

Dear Reviewer,

We appreciate the time and effort you have dedicated providing valuable feedback on the manuscript, and for all the insightful comments. We made changes and the required adjustments in the manuscript.

We hope that we have succeeded in meeting your requirements.

Yours sincerely,

The authors of the manuscript

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

The quality of the manuscript significantly improved. Methods is more detailled and understandable after changes. The more emphasis on the microsatellite and mtDNA results in the Results and Discussion sections are favourable.

Comment to the M&M section: L144: paired samples t-test has to be modified to independent samples t-test

Author Response

Dear Reviewer,

We appreciate the time and effort you have dedicated to providing valuable feedback on the manuscript. We made changes and the required adjustments in the manuscript.

Comments and Suggestions for Authors

The quality of the manuscript significantly improved. Methods is more detailled and understandable after changes. The more emphasis on the microsatellite and mtDNA results in the Results and Discussion sections are favourable.

Comment to the M&M section: L144: paired samples t-test has to be modified to independent samples t-test

Answer: Thanks to the reviewer for the suggestion. The sentence was corrected. (Line 154)

 We hope that we have succeeded in meeting your requirements.

Yours sincerely,

The authors of the manuscript

Author Response File: Author Response.pdf

Reviewer 3 Report

Overall, I am pleased with the modifications the authors have made to their manuscript. There are only a few points that I would like them to correct or edit (see below).

1. The authors still do not mention one inherent issue of using mtDNA to reconstruct genetic relationships: it is only transmitted from mother to offspring, and therefore only recovers maternal lineages. Importantly, mtDNA might be very discordant with nuclear DNA, and can provide a biased perspective of the evolutionary history of the species. Moreover, mtDNA might show an absence of phylogeographic signal, as is the case for the domestic horse. Please, mention these caveats in the manuscript, either in results or in discussion, for the sake of transparency.

2. Figure 5. I appreciate the efforts to highlight the few individuals whose genetic assignment was surprising, but I still do not understand what the X and Y axes stand for. What do the values on the two scales represent? What is the underlying method used for dimensionality reduction? Please be more explicit, for the reader’s convenience.

3. Figure 6. There is still no mention of the meaning of shapes (squares, triangles, diamonds and circles), and colours. Besides, I am wondering why the authors did not compute any quick Maximum-Likelihood phylogenetic inference? I strongly encourage them to do so, as a tree reconstruction with bootstrap support at each node would be much more impactful than this neighbour-joining tree.

Author Response

Dear Reviewer,

We appreciate the time and effort you have dedicated to providing valuable feedback on the manuscript, and for all the insightful comments. We made changes and the required adjustments in the manuscript.

Comments and Suggestions for Authors

1. The authors still do not mention one inherent issue of using mtDNA to reconstruct genetic relationships: it is only transmitted from mother to offspring, and therefore only recovers maternal lineages. Importantly, mtDNA might be very discordant with nuclear DNA, and can provide a biased perspective of the evolutionary history of the species. Moreover, mtDNA might show an absence of phylogeographic signal, as is the case for the domestic horse. Please, mention these caveats in the manuscript, either in results or in discussion, for the sake of transparency.

Answer: Thanks to the reviewer for the suggestion. Done as suggested. Two sentences were added. (Line 683-687)

2. Figure 5. I appreciate the efforts to highlight the few individuals whose genetic assignment was surprising, but I still do not understand what the X and Y axes stand for. What do the values on the two scales represent? What is the underlying method used for dimensionality reduction? Please be more explicit, for the reader’s convenience.

Answer: Thanks for the suggestion. Done as suggested. Figure 5 was corrected and one sentence was added (Figure 5; Line 397-398)

3. Figure 6. There is still no mention of the meaning of shapes (squares, triangles, diamonds and circles), and colours. Besides, I am wondering why the authors did not compute any quick Maximum-Likelihood phylogenetic inference? I strongly encourage them to do so, as a tree reconstruction with bootstrap support at each node would be much more impactful than this neighbour-joining tree.

Answer: Thanks for the suggestion. Done as suggested. Figure 6 was corrected (Neighbour-Joining tree was replaced with Maximum likelihood phylogenetic tree) and one additional sentence was added (Figure 6; Line 452-453)

We hope that we have succeeded in meeting your requirements.

Yours sincerely,

The authors of the manuscript

Author Response File: Author Response.pdf