Validation of Genome-Wide SSR Markers Developed for Genetic Diversity and Population Structure Study in Grain Amaranth (Amaranthus hypochondriacus)

Round 1

Reviewer 1 Report

The authors present an interesting study on a species of economic and health importance. In general, the study is well structured, well presented with an interesting and complete introduction. The methods section needs some supplementation, but the results and discussion sections need a little more work (see specific comments). As English is not my mother language, I am not able to judge this well.

Introduction

Introduction is well written, pertinent, and complete that permit to understand the context of the research.

Line 63. “in the development of superior cultivars for cop improvement”. I think that is “crop” not cop.

Line 97: “Therefore, the present investigation was undertaken to assess the informativeness of the g-SSR markers and screen them against 94 different accessions of Amaranth hypochondriacus to study their molecular diversity and population structure.”. I propose that authors make some precisions about the level of geographic area of the study. For example, how many countries or localities o states etc.…

Methods

 Line 121: “A total of 57 genomic SSR (g-SSR) markers that were developed in-house”. it is necessary to explain from which information and with which program the authors have design the markers.

Line 129: “annealing temperature (determined by gradient PCR for each primer pair) for 40 s”. Here, it will be good to refer to Table S2 for the final annealing temperature of the markers used.

Line 140. The authors determine genetic diversity and heterozygosity. Nevertheless, the genetic diversity corresponds to the expected heterozygosity. By following the program manual used by the authors, they can obtain two types of genetic diversity: the unbiased estimator and the biased estimator. Please check carefully which type of parameter the authors want to use.

Line 152-154: “A total of three runs were carried out for each value of K that was set from 1-10 with burn-in iterations of 100,000, accompanied by 100,000 Markov Chain Monte Carlo (MCMC)”. This used is really minimum for Structure. I propose 10 run for each K value and maybe increase to 500,000 for both. Then, authors could check results for the best K, and if it is 2, check for k=3 to see the relation with the phylogeny. Sometimes, even the best K is 2 (for example), some sub-structure could be observed.

Results

Line 167: “in the present study”. Not necessary.

Line 169: “for all the 94 accessions studied”. Not necessary.

Table 1: I don’t understand what is the start and stop for each marker. Please, clarify.

Table 2: I think it will be interesting integrate genetic diversity parameters for each group, not only polymorphism.

Lines 200-226: It is not necessary to write all the accession numbers in the text because they are in the figure. Text will be clearer without this information and just refer to the figure.

Discussion

 Line 313-315: “Dendrogram of A. hypochondriacus showed three major groups which is consistent with the results of Wang et al., 2013 [39] they also obtained three cluster with different amaranth species”. This does not really make sense for a comparison because we do not know the groups studied and it is normal that some studies also get 3 groups. This is not a valid comparison or justification.

Line 331-332: The first two principal coordinates reported 11.72 % and 8.51 % variation respectively amounting to a total cumulative variation of 20.23%”. Please don’t repeat the results, just refer to the total variation to make the comparison.

General comment: The difference observed between the three population structure analyses is quite disturbing, especially when one sees some logic for the PCoA analysis while the phylogeny and Structure analysis does not present any at all although it marks different clusters but without any apparent explanation. I would tend to think that the PCoA analysis is the most likely, and that methodological adjustments to the other analyses should perhaps be made. The NJ method is probably not the best, and other types of phylogenetic analysis could be tried. A historical search of the seed origin might give some explanation. Also, the AMOVA analysis is based on the 2 clusters proposed by Structure implying that it is the most correct. I suggest rethinking the population analyses in order to be able to present something more coherent with hypotheses on the proposed clusters.

Conclusions

Line 362-363: “These markers distributed the amaranth accessions into different sub-populations effectively”. Maybe, but not consistently.

Line 365: what are MAS and GWAS. Please clarify.

Author Response

Authors would like to thank reviewer for their critical review and very constructive suggestions for improvement of the present manuscript. All the suggestions given has been incorporated in the manuscript. 

Comment No. 1. Introduction is well written, pertinent, and complete that permit to understand the context of the research. Line 63. “in the development of superior cultivars for cop improvement”. I think that is “crop” not cop.

Reply.  Correction has been done in the manuscript as suggested. (Line no. 68).

Comment No. 2. Line 97: “Therefore, the present investigation was undertaken to assess the informativeness of the g-SSR markers and screen them against 94 different accessions of Amaranth hypochondriacus to study their molecular diversity and population structure.”. I propose that authors make some precisions about the level of geographic area of the study. For example, how many countries or localities o states etc.…

Reply. The details have been added in the introduction (Line no. 106-107) as well as in the material and method section (line no. 112-114).

 Comment No. 3. Line 121: “A total of 57 genomic SSR (g-SSR) markers that were developed in-house”. it is necessary to explain from which information and with which program the authors have design the markers.

Reply. The genome sequence data information and software tools used for the generation of g-SSR markers has been incorporated in the materials and methods section of the manuscript (Line no. 132-136).

Comment No. 4. Line 129: “annealing temperature (determined by gradient PCR for each primer pair) for 40 s”. Here, it will be good to refer to Table S2 for the final annealing temperature of the markers used.

Reply. Done as suggested (Line no. 146).

Comment No. 5. Line 140. The authors determine genetic diversity and heterozygosity. Nevertheless, the genetic diversity corresponds to the expected heterozygosity. By following the program manual used by the authors, they can obtain two types of genetic diversity: the unbiased estimator and the biased estimator. Please check carefully which type of parameter the authors want to use.

Reply. We want to use observed heterozygosity and same has been incorporated in the manuscript (Table 1). 

Comment No. 6. Line 152-154: “A total of three runs were carried out for each value of K that was set from 1-10 with burn-in iterations of 100,000, accompanied by 100,000 Markov Chain Monte Carlo (MCMC)”. This used is really minimum for Structure. I propose 10 run for each K value and maybe increase to 500,000 for both. Then, authors could check results for the best K, and if it is 2, check for k=3 to see the relation with the phylogeny. Sometimes, even the best K is 2 (for example), some sub-structure could be observed.

Reply. We performed STRUCTURE analysis again with the parameters suggested by the reviewer, ten runs were for each K value that was set from 1-10 with the burn-in iterations of 500,000. Then, it gave similar results. The peak at k=2 remains unchanged. We also referred to the bar plot at k=3, but no correlation with the phylogeny was observed. 

Comment No. 7. Line 167: “in the present study”. Not necessary

Reply. The string “in the present study” has been removed from the manuscript as suggested (Line no.185).

Comment No. 8. Line 169: “for all the 94 accessions studied”. Not necessary

Reply. The string “for all the 94 accessions studied” has been removed from the manuscript as suggested (Line no. 187).

Comment No. 9. Table 1: I don’t understand what is the start and stop for each marker. Please, clarify.

Reply. The same has been explained in the revised Table 1.

Comment No. 10. Table 2: I think it will be interesting integrate genetic diversity parameters for each group, not only polymorphism.

Reply. Thanks for suggestion. The genetic diversity parameters such as Shannon’s information index (I), Observed Heterozygosity (Ho), and Expected Heterozygosity (He) has been incorporated in the Table 2.

Comment No. 11. Lines 200-226: It is not necessary to write all the accession numbers in the text because they are in the figure. Text will be clearer without this information and just refer to the figure.

 Reply. The accessions have been removed from the Lines 200-226 and Figure 2 is referred at the last (Line no.230-254).

Comment No. 12. Line 313-315: “Dendrogram of A. hypochondriacus showed three major groups which is consistent with the results of Wang et al., 2013 [39] they also obtained three cluster with different amaranth species”. This does not really make sense for a comparison because we do not know the groups studied and it is normal that some studies also get 3 groups. This is not a valid comparison or justification.

Reply. The sentence has been reframed as suggested. (Line no. 359-364)

Comment No. 13. Line 331-332: The first two principal coordinates reported 11.72 % and 8.51 % variation respectively amounting to a total cumulative variation of 20.23%”. Please don’t repeat the results, just refer to the total variation to make the comparison.

Reply. The repetition has been removed from the Line No.331-332 as suggested (Line no 378-379).

Comment No. 14. General comment: The difference observed between the three population structure analyses is quite disturbing, especially when one sees some logic for the PCoA analysis while the phylogeny and Structure analysis does not present any at all although it marks different clusters but without any apparent explanation. I would tend to think that the PCoA analysis is the most likely, and that methodological adjustments to the other analyses should perhaps be made. The NJ method is probably not the best, and other types of phylogenetic analysis could be tried. A historical search of the seed origin might give some explanation. Also, the AMOVA analysis is based on the 2 clusters proposed by Structure implying that it is the most correct. I suggest rethinking the population analyses in order to be able to present something more coherent with hypotheses on the proposed clusters.

Reply. The structure analysis displays availability of two genetically distinct population. However, the PCoA showed relationship between each accession that can be used to understand the introduction of this crop and further diversification nearby geographical area due to its cross-pollination nature. We did try to redraw the dendrogram with different methodology, however all of them showed similar results, hence we have retained the NJ method.

Comment No. 15 Line 362-363: “These markers distributed the amaranth accessions into different sub-populations effectively”. Maybe, but not consistently.

Reply. The correction has been made in the manuscript as suggested. (Line no 400-401)

Comment No. 16. Line 365: what are MAS and GWAS. Please clarify.

Reply. The details have been incorporated in the manuscript (line no. 413-414).

Reviewer 2 Report

This work focuses on the evaluation of the status of genetic diversity and classification of 94 Amaranthus hypochondriacus accessions based on 36 polymorphic genomic SSR (g-SSR). The manuscript is generally well written and structured and the methods are appropriate. However, the manuscript needs some improvements and more clarifications in some sections.

-The authors are invited to avoid using excessive number of references for the same statement: They are highly polymorphic, multi-allelic, co-dominant, highly reproducible, transferable to related species, and easy  detection by polymerase chain reaction (PCR), making them suitable for the under-80 standing genetic diversity in many plant species [23,24,25,26,27,28,29].

- The reproduction pattern of the studied species should be added in the introduction and in the discussion: autogams, allogames ???

-The number of populations and number of accessions per population should be mentioned in plant material.

- The number of seedlings by accession used to extract genomic DNA should be clarified. Did the authors use one single seedling (originated from one seed) by accession to extract DNA??? Generally DNA extraction was made from a mixture of ten to 15 seedlings. Using only one seedling by accession may under estimate the genetic diversity of the studied populations.

-More information about the development of the used SSR molecular markers was needed.

-Please provide the authors reference for Darwin software.

-The redaction of the section 3.2. Phylogenetic Analysis of Amaranth Accessions should be improved: please avoid citing all the genotypes.

-The use of genotype or accessions should be decided according to the real status of the investigated plant material.

- I think that the AMOVA analysis should be performed first based on the studied real geographic populations and then on the two populations as obtained by the model-based population structure analysis.

-The reproduction pattern is among the main factor that explain the status of genetic diversity and genetic structure. This lacks in the discussion.

Author Response

Authors would like to thank reviewer for their critical review and very constructive suggestions for improvement of the present manuscript. All the suggestions given has been incorporated in the manuscript. 

Comment No. 1. The authors are invited to avoid using excessive number of references for the same statement: They are highly polymorphic, multi-allelic, co-dominant, highly reproducible, transferable to related species, and easy detection by polymerase chain reaction (PCR), making them suitable for the understanding genetic diversity in many plant species [23,24,25,26,27,28,29].

Reply. The names of the different plant species in which this type of study has been performed have been incorporated with the reference to avoid the excessive number of references for the same statement (line no. 86-89).

Comment No. 2. The reproduction pattern of the studied species should be added in the introduction and in the discussion: autogams, allogames ???

Reply. The studied species is cross pollinated and it has been incorporated in the revised manuscript in appropriate places. (Line no. 38 &401)

Comment No. 3. The number of populations and number of accessions per population should be mentioned in plant material.

Reply. The number of accessions per population has been incorporated in the plant material section of the manuscript. (Line no. 112-114)

Comment No. 4. The number of seedlings by accession used to extract genomic DNA should be clarified. Did the authors use one single seedling (originated from one seed) by accession to extract DNA??? Generally DNA extraction was made from a mixture of ten to 15 seedlings. Using only one seedling by accession may under estimate the genetic diversity of the studied populations.

Reply. The number of seedlings used per accession for the genomic DNA isolation in the present study has been incorporated in the revised manuscript. (Line no. 122-124)

Comment No. 5. More information about the development of the used SSR molecular markers was needed.

Reply. The information about the development of g-SSR markers used in this study has been updated in the manuscript as suggested. (Line no. 132-136)

Comment No. 6. Please provide the authors reference for Darwin software.

Reply. The author reference no. [46] for Darwin software has been incorporated in the manuscript.

Comment No. 7. The redaction of the section 3.2. Phylogenetic Analysis of Amaranth Accessions should be improved: please avoid citing all the genotypes.

Reply. The accession numbers have been removed from the Section No 3.2 of the manuscript as suggested.

Comment No. 8. The use of genotype or accessions should be decided according to the real status of the investigated plant material.

Reply. The word “genotype” has been replaced by “accession” according to the real status of the investigated plant material as suggested.

Comment No. 9. I think that the AMOVA analysis should be performed first based on the studied real geographic populations and then on the two populations as obtained by the model-based population structure analysis.

Reply. The AMOVA analysis based on the geographic populations and model based AMOVA has been performed and the Figures 5a, 5b, Table S4 as well as the text have been incorporated in the manuscript as suggested.

Comment No. 10. The reproduction pattern is among the main factor that explain the status of genetic diversity and genetic structure. This lacks in the discussion.

Reply. The information about reproduction pattern of the studied species has been incorporated in the discussion section of the manuscript as suggested.

Reviewer 3 Report

Generally, MS deals with a necessary subject, and I do have a few notes addressing the authors:
1 ) Figure 2 is not recommended to be included in MS.
2) Authors may review the paragraph between rows 207-226; No need to include accession names in such detail.
2) The sentences in rows 304 and 305 also need review; how are they not admixture? According to which criteria?

Author Response

Comment No. 1. Figure 2 is not recommended to be included in MS.

Reply: The Figure No. 2 has been moved from main text to Supplementary materials as supplementary Figure S1.

Comment No. 2. Authors may review the paragraph between rows 207-226; No need to include accession names in such detail.

Reply. The accession numbers have been removed from the row no. 207-226 as suggested. (Line no.230-254)

Comment No. 3. The sentences in rows 304 and 305 also need review; how are they not admixture? According to which criteria?

Reply. The criteria of considering genetically pure and admixture accession used in this study has been explained in the result section of the manuscript line No. 264-266.

Round 2

Reviewer 1 Report

For aesthetic reasons, I recommend that the column headings in the tables are presented in abbreviated form and the details are captioned. For example, for Observed heterozygosity put Ho in the column, for Gene diversity put GD ect.... this will lighten the table.

I have no other recommendation

Author Response

Comment No. 1.

For aesthetic reasons, I recommend that the column headings in the tables are presented in abbreviated form and the details are captioned. For example, for Observed heterozygosity put Ho in the column, for Gene diversity put GD ect.... this will lighten the table.

Reply. The column headings in the tables presented in the manuscript have been updated in abbreviated form, and the captions are mentioned in the text as suggested.

Comment No. 2.

I have no other recommendation

Reply. Thank you for your valuable recommendations to improve our manuscript.

Reviewer 2 Report

This manuscript could be accepted for publication.

Author Response

Comment No. 1. This manuscript could be accepted for publication.

Reply. Thank you very much for accepting our manuscript for publication.