Review Reports - Response of African Sorghum Genotypes for Drought Tolerance under Variable Environments

Round 1

Reviewer 1 Report

The work is interesting and aims to evaluate drought tolerance and the effect of genotype-environment interaction (GEI) on grain yield of a population of African sorghum genotypes. The authors seek to identify high-yielding and drought-adapted genotypes for direct production and also for use in breeding programs. The text is well organized, and in all sections the texts are developed with good writing. However, I have some doubts and recommendations that are below:

1) I didn't understand "Wilson]" which is spelled on line 79.

2) On line 86 put a comma after et al. in Ouedraogo et al. 2017; Abreha et al. 2022).

3) do the same in: (Sabadin et al. 2012; Abreha et al. 2022).

4) Check throughout the text and place the comma in situations where they are not placed.

5) In the introduction, lines 113 and 114 contain quotes that refer readers to texts on GGE and AMMI. I suggest placing in parentheses two more citations, which address the AMMI and GGE model with classical and Bayesian analysis. These works bring information and also offer the possibility of new analyses. These situations are current and relevant, including the GGE assessment of sorghum genotypes. The articles are:

4de Oliveira, L. A., da Silva, C. P., da Silva, A. Q., Mendes, C. T. E., Nuvunga, J. J., Nunes, J. A. R., ... & Filho, J. S. D. S. B. (2022). Bayesian GGE model for heteroscedastic multienvironmental trials. Crop Science.

da Silva, C. P., de Oliveira, L. A., Nuvunga, J. J., Pamplona, A. K. A., & Balestre, M. (2015). A Bayesian shrinkage approach for AMMI models. Plos one, 10(7), e0131414.

6) Lines 295 and 296 - Here the productivity in the E1 environment is being described, as I understand it. But notice that only in Table 2 G45, G39, G28, G18, G16 also recorded grain yield > 4.50 t/ha. When looking at Table S2 there are more genotypes, check.

7) Line 300- G129 is written twice.

8) Lines 303 and 304- has repeated information similar to what appears on lines 298 and 299, check it out.

9) Lines 596 and 507 - I verified by the values of the sum of squares of the GEI and the main components that the contributions that are indicated in the text are not consistent. From the calculations I made PC1, PC2 and PC3 correspond, respectively, 61%, 19% and 13%. So the first three axes would explain about 93%. Please check this out. Furthermore, the FR statistic is considered under the null hypothesis that there are no more than n terms determining the interaction. Thus, a significant result by the test suggests that at least one multiplicative term must still be added to the already adjusted n. In their results this implies that AMMI 4 would be the best model and the fourth axis should be adjusted. The AMMI 4 is the winning model in the model selection using the Fr. You didn't make that explicit in the text.

10) Regarding the result of the AMMI analysis (Table 6): I find it even interesting to present the results of the different models, but note that the different models bring different results for genotypes. Which of the models should we base ourselves on? As I said, remember that by the method you used in the selection, the best model is AMMI 4. You used the average predicted by the model in each environment as a selection criterion. In practical works, the authors present the AMMI1 biplot graphic analysis if only the first principal component is significant. Or AMM2 for models where two or more principal axes are significant. Although it is possible to consider biplot with three axes, or the possibility of plotting sequential axes for three-axis mobiles. I don't object to your analysis, but I still think you should indicate the best model based on the method you used.

11) In the item "AMMI Stability Value (ASV)" I believe it would be interesting to plot the AMMI 2 biplot, only with the genotypes and environments that contribute to the interaction. Thus, those considered stable based on the AVS would not be plotted. It could be below Table 8.

12) Dehghani, H., Ebadi, A., and Yousefi, A. (2006) are in the references, but are not cited in the text, check. In fact, check all citations to make sure they are all in the references and vice versa.

Author Response

07 February 2023

Dr. Anita Pizurica

Section Managing Editor,

MDPI Agronomy

Dear Dr. Pizurica

Submission of a revised manuscript ID agronomy-2170137.R1

Thank you for the feedback on our manuscript with ID agronomy-2170137.R1 Entitled: “Response of African sorghum genotypes for drought tolerance under variable environments.” We are grateful for the valuable insights and suggestions of the reviewers, which have further enriched the manuscript.

We have carefully made the required revisions, indicated using track changes and highlighted text in the present submission, as shown in red font text. Please find below the point-by-point response and amendments carried out following the reviewers’ and editor’s comments.

Reviewer’s comments:

Comment 1: I didn’t understand “Wilson]” which is spelled on line 79

Response 1: Thank you very much for taking the time to review our manuscript. We have now carefully paid attention to all the queries and comments raised to improve the manuscript. “Wilson” is the name of the authority’ who had described the species Colletotrichum graminicola. The closed bracket is a mistake and has been removed. The line now read as follows:

e.g., anthracnose caused by Colletotrichum graminicola (Ces.) Wilson [8].

Comment 2: On line 86 put a comma after et al. in Ouedraogo et al. 2017; Abreha et al. 2022).

Response 2: Revised.

Comment 3: do the same in: (Sabadin et al. 2012; Abreha et al. 2022).

Response 3: Revised.

Comment 4: Check throughout the text and place the comma in situations where they are not placed.

Response 4: The entire manuscript has been reviewed, and grammatical errors have been fixed.

Comment 5: In the introduction, lines 113 and 114 contain quotes that refer readers to texts on GGE and AMMI. I suggest placing in parentheses two more citations, which address the AMMI and GGE model with classical and Bayesian analysis. These works bring information and also offer the possibility of new analyses. These situations are current and relevant, including the GGE assessment of sorghum genotypes. The articles are:

da Silva, C. P., de Oliveira, L. A., Nuvunga, J. J., Pamplona, A. K. A., & Balestre, M. (2015). A Bayesian shrinkage approach for AMMI models. Plos one, 10(7), e0131414.

Response 5: Thank you for the useful citations, which are included in-text and reference list.

Comment 6: Lines 295 and 296 - Here the productivity in the E1 environment is being described, as I understand it. But notice that only in Table 2 G45, G39, G28, G18, G16 also recorded grain yield > 4.50 t/ha. When looking at Table S2 there are more genotypes, check.

Response 6: Additional information is provided on high-yielding sorghum genotypes including BLUPs estimates for grain yield in supplemental Table 1. The sentence reads as follows:

Under environment E1, 32 genotypes recorded grain yield > 4.50 t/ha, and the top yielders included G9 (5.41 t/ha), G144 (5.15 t/ha), G123 (4.99 t/ha), G180 (4.93 t/ha), G119 (4.93 t/ha) and G15 (4.87 t/ha) (Table S2).

Comment 7: Line 300- G129 is written twice.

Response 7: Removed.

Comment 8: Lines 303 and 304- has repeated information similar to what appears on lines 298 and 299, check it out.

Response 8: Repeated information has been removed.

Comment 9a: Lines 596 and 507 - I verified by the values of the sum of squares of the GEI and the main components that the contributions that are indicated in the text are not consistent. From the calculations I made PC1, PC2 and PC3 correspond, respectively, 61%, 19% and 13%. So the first three axes would explain about 93%. Please check this out.

Response 9a: Thank you, it was an oversight. The correct values in Table 6 and in-text as follows:

IPCA1, IPCA2 and IPCA3 explained 61.4%, 19.4% and 12.8% of the GEI variation, respectively. The three IPCAs cumulatively explained 93.6% of the total variation.

Comment 9b: Furthermore, the FR statistic is considered under the null hypothesis that there are no more than n terms determining the interaction. Thus, a significant result by the test suggests that at least one multiplicative term must still be added to the already adjusted n. In their results this implies that AMMI 4 would be the best model and the fourth axis should be adjusted. The AMMI 4 is the winning model in the model selection using the Fr. You didn't make that explicit in the text.

Response 9b: Thank you for the useful insight. We are in agreementand modified the text as follows:

The FR statistic is considered under the null hypothesis that no more than n terms determine the interaction. Thus, a significant result by the test suggests that at least one multiplicative term must still be added to the already adjusted n. In the present study, three significant IPCAs were detected, indicating the need for adding a multiplicative term. As a result, AMMI model 4 is the best fitting model for the yield dataset based on the FR test.

Comment 10: Regarding the result of the AMMI analysis (Table 6): I find it even interesting to present the results of the different models, but note that the different models bring different results for genotypes. Which of the models should we base ourselves on? As I said, remember that by the method you used in the selection, the best model is AMMI 4. You used the average predicted by the model in each environment as a selection criterion. In practical works, the authors present the AMMI1 biplot graphic analysis if only the first principal component is significant. Or AMM2 for models where two or more principal axes are significant. Although it is possible to consider biplot with three axes, or the possibility of plotting sequential axes for three-axis mobiles. I don't object to your analysis, but I still think you should indicate the best model based on the method you used.

Response 10: In line with comment 9, we have based the selection of suited genotypes for specific environments based on AMMI model 4. We have now captured these in-text, abstract and conclusions sections.

Comment 11: In the item "AMMI Stability Value (ASV)" I believe it would be interesting to plot the AMMI 2 biplot, only with the genotypes and environments that contribute to the interaction. Thus, those considered stable based on the AVS would not be plotted. It could be below Table 8.

Response 11: We attempted to draw biplots using metan (multi-environment trial analysis) procedure in R version 4.2.0. However, there were too many overlaps when plotting the genotypes. Hence, we have opted to exclude the plots, given that we cannot find a clear distinction to make plausible conclusions.

Comment 12: Dehghani, H., Ebadi, A., and Yousefi, A. (2006) are in the references, but are not cited in the text, check. In fact, check all citations to make sure they are all in the references and vice versa.

Response 12: Thank you for this observation. References cited in-text and now included in the reference list and vice versa.

With best regards

Muhammad Ahmad Yahaya

Corresponding author

Reviewer 2 Report

The authors reported a study on the screening of more than 200 genotypes of Sorghum bicolor to establish the most resistant and drought-sensitive plants. The authors have well framed the important role of sorghum for both human and animal nutrition. Furthermore, as African countries are major sorghum producers, the economic aspects have also been emphasised. Therefore it is important for populations to establish the best genotypes capable of withstanding drought.

From an experimental point of view, the authors subjected the plants to drought stress before anthesis, after anthesis, both in greenhouse and open field crops. Therefore, during the growth they applied different parameters of agronomic efficiency of the crops. Subsequently, after an in-depth statistical analysis, the best and most resistant genotypes to a sub-Saharan climate were established

The manuscript is clear and very precise. I suggest discussing and supplementing the references with the manuscript below which discusses some aspects related to sorghum cultivation:

Rad, S.V.; et al.. Quantitative and Qualitative Evaluation of Sorghum bicolor L. under Intercropping with Legumes and Different Weed Control Methods. Horticulturae 2020, 6, 78. doi: 10.3390/horticulturae6040078

Author Response

07 February 2023

Dr. Anita Pizurica

Section Managing Editor,

MDPI Agronomy

Dear Dr. Pizurica

Submission of a revised manuscript ID agronomy-2170137.R1

Thank you for the feedback on our manuscript with ID agronomy-2170137.R1 Entitled: “Response of African sorghum genotypes for drought tolerance under variable environments”. We are grateful for the valuable insights and suggestions of the reviewers, which have further enriched the manuscript.

Reviewer’s comments:

Comment 1: The authors reported a study on the screening of more than 200 genotypes of Sorghum bicolor to establish the most resistant and drought-sensitive plants. The authors have well framed the important role of sorghum for both human and animal nutrition. Furthermore, as African countries are major sorghum producers, the economic aspects have also been emphasized. Therefore, it is important for populations to establish the best genotypes capable of withstanding drought.

Response 1: Thank you very much for taking the time to review our manuscript. We have now carefully paid due attention to all the queries and comments raised and improved the manuscript.

Comment 2: The manuscript is clear and very precise. I suggest discussing and supplementing the references with the manuscript below which discusses some aspects related to sorghum cultivation:

Response : The suggested reference has been added to the manuscript.

With best regards

Muhammad Ahmad Yahaya

Corresponding author