Multidimensional Analysis of Diversity in Genotypes of Winter Oilseed Rape (Brassica napus L.)
Round 1
Reviewer 1 Report
Observations:
1) Rewrite or precise the objective indicated in the Introduction is few descriptive.
2) In subsection 2.2. Experimental conditions were stated that the experimental design and conditions such as a randomized block design with four replicates and two locations of evaluation. However, in the Table 5 present results of a completely randomized design because is absent the source of variation blocks which should be nested in locations and years but also with 2 df for locations when is 2-1=1 and years have similar confusion (it is no clear the years definition). In consequence, the ANOVA does not correspond to the experimental stated and probably all inferences can change.
3) All arguments in the Introduction, Results and Discussion were addressed to mark a contribution for formulation of plan breeding strategies in oilseed rape. However, along work authors did not present information related with that objective or is limited. In this sense, the recommendations are:
- Estimate the genotypic correlation across environments and no phenotypic ones such as is indicated in the section 2. Correlation analysis, because we need information on genetic relationships.
- Based on a combined ANOVA estimate the components of variance as genotypic (G), environmental (E) and GE across traits evaluated. This estimates are useful to measure or compare the genetic effect.
- In this type of works is important obtain estimates of heritability and quantify the environmental effects or repeatability (nongenetic effects). In this case, The GGE biplot method is very useful to study the GE interaction relationship among locations.
3) In the section of Results there is not information on genotypes across environments such as means of yield or another trait to each environment. Therefore, any reader can see the behavior variation of genotypes across environments.
4) In the subsection 3.4. Multivariate Regression Analysis [MRA), all r2 indicated in the Table 7 are less than 0.6, then the prediction to each environment is limited or very low and will be recommendable to do a new MRA using all information available.
Author Response
Response to Reviewer 1 Comments
Point 1: Rewrite or precise the objective indicated in the Introduction is few descriptive.
Response: According to Reviewer’s advice the main objective on this study has been precised.
Point 2: In subsection 2.2. Experimental conditions were stated that the experimental design and conditions such as a randomized block design with four replicates and two locations of evaluation. However, in the Table 5 present results of a completely randomized design because is absent the source of variation blocks which should be nested in locations and years but also with 2 df for locations when is 2-1=1 and years have similar confusion (it is no clear the years definition). In consequence, the ANOVA does not correspond to the experimental stated and probably all inferences can change.
Response: In Table 5 we added the row with results for source of variation block. We corrected Table 5. Previously, we mixed up names year and location by accident. Now is correct.
All arguments in the Introduction, Results and Discussion were addressed to mark a contribution for formulation of plan breeding strategies in oilseed rape. However, along work authors did not present information related with that objective or is limited. In this sense, the recommendations are:
Point 3: Estimate the genotypic correlation across environments and no phenotypic ones such as is indicated in the section 2. Correlation analysis, because we need information on genetic relationships.
Response: We added genotypic correlations. We added additional figure (Figure 2) and text: “The results of the genotypic correlation analysis are shown in Figure 2. BF was significantly negatively correlated with SY (-0.76), NS (-0.51) and WTS (-0.52). Positive genotypic correlations were observed between: SY and NS (0.72), SY and WTS (0.56), LS and NS (0.66) as well as LS and WTS (0.58) (Figure 2).”.
Point 4: Based on a combined ANOVA estimate the components of variance as genotypic (G), environmental (E) and GE across traits evaluated. This estimates are useful to measure or compare the genetic effect.
Response: The residual maximum likelihood (REML) was used to estimate the variance components. The likelihood ratio test was used to test significance for each term. The variance components were used also to estimate broad sense heritability for each trait.
Point 5: In this type of works is important obtain estimates of heritability and quantify the environmental effects or repeatability (nongenetic effects). In this case, The GGE biplot method is very useful to study the GE interaction relationship among locations.
Response: We estimated the heritability in broad sense and these results added in the manuscript. We added a tri-plot as a combine of a genotype × environment biplot and a genotype × trait biplot by using the same set of genotype scores.
Point 6: In the section of Results there is not information on genotypes across environments such as means of yield or another trait to each environment. Therefore, any reader can see the behavior variation of genotypes across environments.
Response: We added five tables with mean values and standard deviations for all genotypes in particular environments.
Point 7: In the subsection 3.4. Multivariate Regression Analysis [MRA), all r2 indicated in the Table 7 are less than 0.6, then the prediction to each environment is limited or very low and will be recommendable to do a new MRA using all information available.
Response: We added multivariate regression analysis for all six environments jointly. We added one column in Table 7 and description of these results: “Across environments all four traits statistically significant affected seed yield: beginning of flowering and length of silique have a negative effects, however the number of seeds per silique and weight of 1000 seed – positive effects (Table 7). Coefficient of determination ranged from 27.0% (in E6) to 59.3% (in E3). However for multivariate regression model using all information available across environments coefficient of determination was equal to 73.6% (Table 7).”.
Reviewer 2 Report
The manuscript entitled "Multidimensional Analysis of Diversity in Genotypes of Winter Oilseed Rape (Brassica napus L.)" by Jan Bocianowski and Alina Liersch is very well written and described potential role of winter rapeseed in breeding programme. The authors designed the experiments very carefully and performed the statistical analysis. Mostly they have collected the genotypes from European origin and that is the major drawback. They could have add some more geological locations. Manuscript is well written but it needed some corrections in terms of grammars. I found authors could highlight their findings more precisely in conclusion. I suggest to rewrite the conclusion.
Author Response
Response to Reviewer 2 Comments
Point 1: The manuscript entitled "Multidimensional Analysis of Diversity in Genotypes of Winter Oilseed Rape (Brassica napus L.)" by Jan Bocianowski and Alina Liersch is very well written and described potential role of winter rapeseed in breeding programme.
Response: Thank you very much.
Point 2: The authors designed the experiments very carefully and performed the statistical analysis.
Response: Thank you very much.
Point 3: Mostly they have collected the genotypes from European origin and that is the major drawback. They could have add some more geological locations.
Response: According to Reviewer׳s advice in our future research we will plan to test not only European cultivar and breeding materials but also some Asian winter oilseed rape genotypes and varieties from China, Korea or Japan (for example from OMGC – Oil Crops Middle-term Genebank of China).
Point 4: Manuscript is well written but it needed some corrections in terms of grammars.
Response: English was revised.
Point 5: I found authors could highlight their findings more precisely in conclusion. I suggest to rewrite the conclusion.
Response: The conclusions have been rewritten to outline the mains trends emerging from the results, including the usefulness of the applied of statistical method to analyse genotypes, environments and GEI and to determine particular genotypes, which will be probably used as starting materials in future breeding programs (HO, HOLL or RS).
