Genetic Diversity and Population Structure of Brachiaria (syn. Urochloa) Ecotypes from Uganda
, Swidiq Mugerwa 1, Martina Kyalo 2, Collins Mutai 2, Robert Mwesigwa 3, Appolinaire Djikeng 2,†and Sita Ghimire 2,*Round 1
Reviewer 1 Report
Authors try to analyze the genomic diversity of Brachiaria collection from Uganda. The authors collected 99 accessions from the different ecological regions in Uganda and genotyped them using 24 SSR markers.
1) In the MS, the authors identified 1122 alleles for 24 SSR markers for 99 accessions. ~48 alleles for each marker, this is too high.
2) In the MS, the authors showed the results in a way that data showed, but the biological question authors want to answer is not clear. The authors may want to extract the take-home information from the results.
Author Response
Reviewer 1
Comment 1: In the MS, the authors identified 1122 alleles for 24 SSR markers for 99 accessions. ~48 alleles for each marker, this is too high.
Response: Agree. We revisited raw SSR data, reanalyzed using R software and the number of alleles for each SSR marker was counted. Indeed, we found that the power marker had overestimated the allele number by counting the allele combinations for each locus rather than counting allele. Now using R software, we have come up with a total of 584 allele (average of 24.3 per allele). While revisiting SSR data we found that each marker/locus yielded multiple alleles ranging between 4 to 12 alleles as expected for polyploid species. Revised data on allele number for each marker is now presented in Table 1 (line 145) and also in abstract (line 27) and result section.
2) In the MS, the authors showed the results in a way that data showed, but the biological question authors want to answer is not clear. The authors may want to extract the take-home information from the results.
Response: Agree. The take home messages have been provided in the discussion section (lines 353 to 430) as well as in the conclusion section. These issues have been reflected in the discussion section (433 to 446).
Others
In addition, content and reference are added in the introduction materials and methods and results and conclusion sections have been revised. These can be seen in track change mode.
Reviewer 2 Report
The study reports genetic diversity on 99 Brachiaria clones collected presumably native stands from different regions of Uganda, with the objectiver to establish a nationwide collection.
The overall rationale for the study is good and the Introduction is generally well written.
Critique:
A strong rationale of this study was related to conservation of natural germplasm from natural Brachiaria populations that are threatened by introduction of new plants and modern farming practice. The authors claim "we successfully established the first nation-wide collection of Brachiaria ecotypes in Uganda..." and conclude that plant materials characterized in this study "will form a basis for Brachiaria grass improvement...in Uganda and beyond." This sounds very good, but I feel like it would be helpful if the authors could describe in more detail exactly how this germplasm will be maintained and possibly distributed, for future use, in Uganda and possibly beyond. Will the germplasm be maintained clonally or be seed? Does Uganda have a germplasm repository for this purpose? If not, are there any other options?
As an outside reviewer that is not familiar with the geography of natural history of Uganda, comparisons of genetic diversity within and among the five regions (using AMOVA) seems rather arbitrary. In fact, further analyses of the data (PCA and Structure) do not support these arbitrary divisions, although it was shown that there were some "private alleles" in these groups. The authors did not adequately explain, in the Introduction, why they thought these regions were important. Although there were private alleles in each region, I am not really sure what that means given the rather sparse nature of the sampling. However, it would be reasonable to speculate that materials from these different regions may have alleles conferring unique adaptations if the authors can adequately describe environmental differences among these regions in the Introduction.
The authors did not adequately discuss the main findings (3 genetic groups). Why are there three genetic groups? Could these be different species? How were these three groups distributed across the sampling sites? How were the K=3, K=4, and K=6 groups distributed across the landscape? Why did the authors not show results for K=5? It seems to me that if the authors are interested in five geographic regions, then it is most logical to show results of the K=5 model test!
Line 300: "as predicted by Evanno et al." It sounds like Evanno already analyzed Brachiaria, which is not true. Suggest "based on comparisons of model probabilities as described by Evanno et al."
Table 2: How can you have MAF (minor allele frequency) greater than 0.50? This does not make sense.
Overall, I would say this is borderline between major and minor revisions. I don't think it should take a great deal of effort to address these comments, but I think there are some important opportunities to make the manuscript better. Some things need to be added in the Introduction, Discussion, and possibly the Results (K=5 Structure results) so it is not just a simple matter of text editing.
Author Response
Reviewer 2
Critique
Part 1: A strong rationale of this study was related to conservation of natural germplasm from natural Brachiaria populations that are threatened by introduction of new plants and modern farming practice. The authors claim "we successfully established the first nation-wide collection of Brachiaria ecotypes in Uganda..." and conclude that plant materials characterized in this study "will form a basis for Brachiaria grass improvement...in Uganda and beyond." This sounds very good, but I feel like it would be helpful if the authors could describe in more detail exactly how this germplasm will be maintained and possibly distributed, for future use, in Uganda and possibly beyond. Will the germplasm be maintained clonally or be seed? Does Uganda have a germplasm repository for this purpose? If not, are there any other options?
Response: Each ecotype collected in this study has been maintained as single plant in a vegetative field genebank at NaLIRRI Research Farm at Tororo, Uganda. NaLIRRI, an institution under National Agricultural Research Organization (NARO) is owner of these ecotypes. These germplasms are accessible to other researchers following the Ugandan government’s guidelines for accessing genetic resources and benefit sharing (lines 108-111 and 435-437).
As an outside reviewer that is not familiar with the geography of natural history of Uganda, comparisons of genetic diversity within and among the five regions (using AMOVA) seems rather arbitrary. In fact, further analyses of the data (PCA and Structure) do not support these arbitrary divisions, although it was shown that there were some "private alleles" in these groups. The authors did not adequately explain, in the Introduction, why they thought these regions were important. Although there were private alleles in each region, I am not really sure what that means given the rather sparse nature of the sampling. However, it would be reasonable to speculate that materials from these different regions may have alleles conferring unique adaptations if the authors can adequately describe environmental differences among these regions in the Introduction.
Response: Thank you. The information about Uganda and floral diversity and agroecological zone in Uganda (AEZ) has been provided in introduction section of revised version (lines 80-84). Relationship between districts, regions and AEZ has been provided in material and method section (lines 105 -108). We have also provided information on altitude and rainfall data for each region (lines 327-381). Association of these two parameters with allele frequency has been presented. We have also and have discussed on the importance of private alleles detected in regional population in breeding for adaptation and other traits.
The authors did not adequately discuss the main findings (3 genetic groups). Why are there three genetic groups? Could these be different species? How were these three groups distributed across the sampling sites? How were the K=3, K=4, and K=6 groups distributed across the landscape? Why did the authors not show results for K=5? It seems to me that if the authors are interested in five geographic regions, then it is most logical to show results of the K=5 model test!
Response: Thank you very much for this question. Indeed, as indicated in material and method section (lines 179-180) 10 independent runs of STRUCTURE were performed by setting K from 1–10 with 15 replicates for each K and the software predicted K=3 as most likely number of clusters. Therefore K=4, and K=6 were not discussed. We have provided Figure S1: Plot of changes in ΔK value with the number of subpopulations in page 14 (lines 645- 657) and Figure S1 has been also cited in the results section (lines 308-310).
Line 300: "as predicted by Evanno et al." It sounds like Evanno already analyzed Brachiaria, which is not true. Suggest "based on comparisons of model probabilities as described by Evanno et al."
Response: Thank you – suggestion is well taken into consideration and revised accordingly (line 310).
Table 2: How can you have MAF (minor allele frequency) greater than 0.50? This does not make sense.
Response: Each SSR markers used in this study detected more than one allele in 99 ecotypes. The range of alleles detected for each locus ranged between 4 to 12. Since the Brachiaria grass is polyploid the accurate estimation of allelic dose is not possible and studies on polyploidy usually do not determine the allelic dose (refer K. Saltonstall, Molecular Ecology 2003 12: 1689-1202). Therefore, we opted not to present the allelic dose data. Accordingly, in this revised version we have removed Table 2 and moved PIC and Na values from Table 2 to Table 1.
Overall, I would say this is borderline between major and minor revisions. I don't think it should take a great deal of effort to address these comments, but I think there are some important opportunities to make the manuscript better. Some things need to be added in the Introduction, Discussion, and possibly the Results (K=5 Structure results) so it is not just a simple matter of text editing.
Response: Thank you. This issue has been now addressed.
Others
In addition, content and reference are added in the introduction and materials and methods sections. Some changes are made in the conclusion section as requested. These changes can be seen in relevant sections in track change mode.
Round 2
Reviewer 1 Report
I do not comments at this versions.
Reviewer 2 Report
The article entitled "Genetic Diversity and Population Structure of Brachiaria (syn. Urochloa) Ecotypes from Uganda" provides compelling evidence for regionally significant genetic variation among native Bracharia ecotypes of Uganda.
