Genetic Structure and Historical Dynamics of Pinus densiflora Siebold & Zucc. Populations
Round 1
Reviewer 1 Report (Previous Reviewer 2)
I am reviewing the submitted manuscript for the second time (previous ID: 1945197). In the last review I pointed out a few comments, and most of them were addressed by the authors in the present version of the manuscript. The tables were changed and I find them more suitable now. Therefore, I can recommend the manuscript for publication after just a few minor changes:
Figure 1. – This figure is the same as in the first version of the manuscript. The location and color used to write population names make some of them unreadable.
Figure 3. – In the previous review, I suggested that the clusters visible in the PCA analysis are not so evident. It would not be possible to distinguish them without colors, so it is rather arbitrary. There is certainly some divergence, but also admixture, which was seen mostly in the third diagram (removed by the authors). Anyway, the authors did not refer to this admixture, either in the Results or in the Discussion.
Section 2.5 – Why the maximum number of regions was set to four (line 146)?
Minor spelling mistakes:
- line 68: populations East Asia à populations in East Asia
- line 114: in dividuals à individuals
- line 146: regions set to four à regions was set to four
Author Response
Response to Reviewer 1:
Response:
Thanks very much for reviewing our manuscript! We are very grateful to you for reviewing the paper so carefully. The comments you raised are very helpful for improving the quality of our manuscript! We have carefully considered the suggestion and made some changes.
Figure 1. – This figure is the same as in the first version of the manuscript. The location and color used to write population names make some of them unreadable.
Response:
We have re-drawn and replaced it with a clear version (Page 3, Line 88).
Figure 3. – In the previous review, I suggested that the clusters visible in the PCA analysis are not so evident. It would not be possible to distinguish them without colors, so it is rather arbitrary. There is certainly some divergence, but also admixture, which was seen mostly in the third diagram (removed by the authors). Anyway, the authors did not refer to this admixture, either in the Results or in the Discussion
Response:
We have revised it and added a description of the admixture of individuals among subpopulations in the PCA analysis in the results section (Page 5, line 167-170).
Section 2.5 – Why the maximum number of regions was set to four (line 146)?
Response:
We re-checked the RASP manual and found that the maximum number of regions was no less than the maximum number of unit areas of species. Then we set the maximum number of regions was six and ran. The results based on the new settings were consistent with the initial results. So, to avoid ambiguity, we revised this maximum number of regions to six. We are very sorry for our incorrect (Page 5, Line 149).
Minor spelling mistakes:
line 68: populations East Asia à populations in East Asia
Response:
Yes, we have revised it (Page 2, Line 68).
line 114: in dividuals à individuals
Response:
Yes, we have revised it (Page 4, Line 116).
line 146: regions set to four à regions was set to four
Response:
Yes, we have revised it (Page 5, Line 149).
Author Response File: 
Author Response.docx
Reviewer 2 Report (Previous Reviewer 1)
I thank the Authors for accepting my suggestions and correcting the manuscript accordingly. Even though I still think that some of the results could have been more deeply discussed, I am satisfied with this revision. Before pubblication, there is however one single point for which, as a population geneticist, I do not agree with the Authors' answer, namely, the significance of FST values. In fact, “Since the degree of genetic differentiation among Pinus densiflora populations was not high, with FST at 0.05-0.1 indicating reaching a moderate degree of differentiation, we focused on describing reaching a moderate degree of differentiation and did not consider whether the FST value was significantly different from zero” is not acceptable as an answer.
If a statistical test is not run, one cannot be sure whether the FST values found are due to stochastic fluctuations or to a real differentiation between populations. This way, the information given by FST becomes also of importance for the discussion. According to the software used, a test of significance (a simple permutation test would be enough) can be obtained just one click ahead!
Last, another very minor point: I daresay that (line 372) P. jezoensis did not play an essential role in the formation of Japanese frogs, rather it played a role in their colonization and evolution.
Author Response
Response to Reviewer 2:
Response:
Thanks very much for reviewing our manuscript! We are very grateful to you for reviewing the paper so carefully. The comments you raised are very helpful for improving the quality of our manuscript! We have carefully considered the suggestion and made some changes.
Before publication, there is however one single point for which, as a population geneticist, I do not agree with the Authors' answer, namely, the significance of FST values. In fact, “Since the degree of genetic differentiation among Pinus densiflora populations was not high, with FST at 0.05-0.1 indicating reaching a moderate degree of differentiation, we focused on describing reaching a moderate degree of differentiation and did not consider whether the FST value was significantly different from zero” is not acceptable as an answer.
If a statistical test is not run, one cannot be sure whether the FST values found are due to stochastic fluctuations or to a real differentiation between populations. This way, the information given by FST becomes also of importance for the discussion. According to the software used, a test of significance (a simple permutation test would be enough) can be obtained just one click ahead!
Response:
Yes, we have recalculated Pairwise FST between populations and subpopulations using the software Arlequin and performed significance tests. I have added the new results in Supplementary Table 1. (Page 4, Line 109; Page 5, Line 187-189; Page 8-9, Line 208-209; Supplementary Table 1).
Last, another very minor point: I daresay that (line 372) P. jezoensis did not play an essential role in the formation of Japanese frogs, rather it played a role in their colonization and evolution.
Response:
Yes, we have revised it (Page 12, line 329).
Author Response File: Author Response.docx
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The paper by Zhiyuan and coll. deals with the population genetics of Pinus densiflora, an important Asian conifer whose current distribution covers Eastern China, Korea and Japan. To this end, they utilized 8,376 transcriptome-derived SNPs to assess genetic diversity, differentiation and to make inference about the demographic processes underwent by this species.
The work has merits, mainly because it is the first one to deal with P. densiflora, but in my opinion needs some revision to be more informative about the methods used and the results obtained. Also, in many places the style is not as rigorous as it should be for a scientific paper.
First, I wish the Authors could state why transcriptome-derived SNPs were chosen, instead of genomic SNPs. This choice also impacts on the discussion of the results, because nowhere in the ms. the fact that the SNPs used are from coding regions is stressed. Are there footprints of selection? Are demographic inferences affected in some ways by this?
Second, in my opinion the Materials & Method section is not as informative as it could have been, because of the following.
Table 1 could be safely moved to SM, because the information provided is much the same than the one in Fig. 1, more easily understood.
“The optimal K was determined using Structure Harvester [40] delta K statistics” is wrong, the DK index is by Evanno and coll. (Evanno et al., 2004), Harvester is just a (very) useful online tool.
The long-winded explanation of PCA starting at line 113 is not necessary, just state what the original variables are and how PCs were obtained. In the same way, adjust the legend to Fig. 3, by stating upon which variables PC1, PC2 and PC3 are based.
At line 121, “Neighbor-Joining (NJ) trees were constructed”: please state which was the variable used to obtain the distance matrix. Correct correspondingly the legend to Fig. 4.
The citation for RASP is missing both in the text and in the references (Yu et al., 2015). A brief description of the different models tested and a more detailed description of the chosen one could be useful. Besides, no legend is provided for Table S2. (Line 144: “F represented the Japanese archipelago” actually is “E represents …”).
Later on, at line 166, it is said that “the two subpopulations (CH and JP) analysed using the first two methods were not well represented in the phylogenetic tree”: can this have introduced a bias in the analyses?
In the legend to Fig. 2, what is the meaning of “SL, JH, CH, JP and JK are the subpopulations at K=5, respectively”?
Figure 5 is very difficult to read because of low resolution and I think that something is missing.
The tree in Fig. 6 has very low resolution too, so that its description in paragraph 3.4 is again very difficult to follow. Can you provide a better graphic?
Table S1 shows that FST values are very low among all populations tested, however, are there any values significantly different from zero? This information is not given and the same holds true for the “regional” FST values shown in Tab. 4.
The Discussion is in my opinion quite poor. Given the huge amount of data available, the Discussion should encompass more literature, and discuss more topics (see the suggestion above about the use of transcriptome, but also the reconstruction of ancestral states could be discussed with emphasis on, say, what is known about the climate changes occurred in the reconstructed period of time). As it is, the Discussion is just something like a recollection of some of the results.
Last, please have your English checked.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Reviewer 2 Report
The manuscript “Genetic structure and historical dynamics of Pinus densiflora populations” aimed to study the genetic variation and structure of P. densiflora in East Asia. The authors also wanted to estimate the divergence time of P. densiflora and reconstruct its migration patterns.
In my opinion, the manuscript is interesting and relevant to the field of tree genetics. It certainly gives new information to the previous studies about the genetic variation of P. densiflora, especially concerning its putative migration routes and divergence time. It will be interesting to the broad readership of Forests. The methodology is scientifically sound, I have no concerns about English, but I believe that certain parts of the manuscript need to be improved. Furthermore, the presentation of the results should have been better thought over. Also, in my opinion, some important information is missing in the introduction and discussion of the paper. Below, I specify my comments, which I would like to be addressed by the authors.
Abstract
- Line 20: million years AGO.
- Line 20: The authors wrote about the three large subpopulations – I think that it may be misleading (are they countries? – China, Korea, Japan or SL, CH, and JK groups as in Fig. 2?). I guess it is better to stick to five groups.
Introduction
- Lines 51-53: In my opinion, it is important to remember that transcriptome SNPs are in coding regions. This means that they are not selectively neutral, which may be the problem considering their use to make inferences about P. densiflora population history. Thus, the interpretation of data should be made with caution. I think that the authors should keep it in mind and emphasize it both in the introduction and discussion sections. The same applies to the number of trees analyzed from each population – as it was very low in some cases.
- I am missing a little more information about P. densiflora, i.e. its ecology, present and past situation considering human disturbances, etc.
- Line 67: The authors cited a few past papers about the genetic variation of P. densiflora, but one of them (no. 25 in the references) is about old planted populations of P. thunberghii.
Table 1.
- I think it is not necessary to group populations into provinces, I would rather stick to five main regions and order them in the same way as in section 2.5. In the current version, it gives the impression that the analysis will be carried out using populations grouped into 10 regions (but maybe it is the reason for K ranging from one to ten in Structure).
Table 2.
- Line 197: nucleoid à nucleotide.
- It would be better to order populations as in Table 1.
Figure 1.
- The caption to the TFZ population is not visible.
- I suggest adding the putative migration routes and divergence time to the map.
Figure 5.
- This figure has too low resolution and it is hard to see differences among colors, especially purple vs. red-purple.
- + Figure 6: As for me, it would be easier to interpret if I had the information on which population every tree belongs to.
Materials and methods
- Section 2.5: Why the maximum number of regions was set to four (line 146)?
- Can the authors illustrate what the six tested models looked like?
Results
- Line 162: GTJ should be changed to KNO.
- Line 164: To me, the clusters visible in the PCA analysis are not so evident. It would not be possible to distinguish them without colors, so it is rather arbitrary. I see some divergence, but there is a lot of admixture here.
- It would be better to round FST values to three or four decimal places.
- Section 3.4: This section was a bit too hard for me to follow, but I am not familiar with this type of analysis. Anyway, it could be changed a bit to make it less complicated.
Discussion
- There are a lot of results but little discussion, especially in the 4.1 and 4.2 sections. The discussion concentrates mostly on divergence time and migration pathways. I am also missing some discussion about how range fragmentation, as well as human activities and changes in land use, have affected the genetic variation and structure of P. densiflora populations.
- Can the authors explain why AT and KNO populations differ from the remaining populations? Taking into account their geographic location, it is especially hard to guess why AT constituted a separate genetic cluster. I can see a certain level of admixture, but can the authors think about any specific reason behind it? Can AT be the main source of migrants?
- There are a lot bigger differences in the population HO than HE, with HO exceeding HE, but this greater heterozygosity is not discussed by the authors.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
