Evaluation of Water Quality of Buritis Lake

Round 1

Reviewer 1 Report

Abstract

“Statistical tests were applied to verify existing clusters and correlations between parameters” -> Since the authors did not perform cluster analysis, this information cannot be included in the abstract.

 

Statistical analysis

“For analysis of normality, the Shappiro-Wilk and Kolmogorov-Smirnov tests were performed according to the sample size; the normality tests were associated with the evaluation of measures of central tendency and graphical distributions. For acceptance of the significance of the normality tests, the significance limit of 5% was considered, thus, when the tests presented p-value ≤ 0.05, the data were considered non-parametric and, when they presented p-value > 0.05, they were considered parametric data.” -> The information contained in this paragraph is correct, but in grammatical terms the sentence has to be revised, because in addition to being too long, it is quite confusing.

 

“Subsequently, the ANOVA tukey test for parametric data” -> Authors have to review this information as there is no test going by " the ANOVA tukey test".

 

“For the analysis of the A. cepa test, was used ANOVA, through the Minitab 17 Statistical Software Program, followed by the Tukey or Games-Howell test, the differences were considered significant when p < 0.05.” -> Authors have to review this information as there is no test going by " the A. cepa test". On the other hand, there has to be coherence in the document (which fails a lot), because authors cannot sometimes put “p < 0.05” and other “p-value ≤ 0.05”. This situation appears continuously throughout the document.

 

“PCA (Principal Component Analysis) with centralization of the mediaand Pearson correlation was applied to verify the relationships of the two results obtained two physical-chemical parameters and test of A. cepa, thus finding groups that have similar behaviors or correlations, in this case, was used STATISTICA software version 10.0 (Statsoft, Tulsa, Ok, USA).” -> Authors should correct the sentence as it should be: “Principal Component Analysis (PCA)”; On the other hand, there is a lack of space in “mediaand”. Additionally, authors should clarify what they mean by “and test of A. cepa”, because it is not understood what the authors will test and/or analyse.

 

Results

The entire way of presenting results has to be extensively reviewed, because in addition to inconsistency in the presentation, there are many sentences that are not understood. I describe some examples:

“The root growth of A. cepa roots, submitted to the treatment of the first collection (July, 2017), showed a significant difference (p = 0.001), indicating cytotoxicity of the analyzed samples. It should be noted that the post hoc analysis (Games - Howell test), of the treatment, showed a significant difference only for the sample at point P6 (p<0.05).” (…) “The statistical analysis of the root growth of the meristematic cells of A. cepa, from the collection carried out in October 2017, demonstrated that there was no significant difference (p = 0.207), between the samples and the negative control. The result of the MI of the 2nd collection, confirms the results of cytotoxicity, based on root growth (p = 0.07), between the negative control and the samples of P1, P2, P3, P4, P5 and P6.” -> clear evidence of lack of consistency in the p-value presentation

 

“The post hoc test (p<0.05) indicated the significant difference of the P2 point in relation to the negative control. The MI indicated a significant difference between the negative control and the P2 point (p = 0.008).” -> idem

“The post hoc analysis (Games-Howell test), of the treatment, showed a significant difference for the sample of point P2 (p=1.164).” -> it is still surprising to present a p-value greater than 1, because given that this value is a probability, then what is here does not make sense.

Still in the presentation of the results, the authors sometimes identify the multiple comparison tests used, but other times they do not.

On the other hand, it is also not clear where the results from the application of the t-test and the Wilcoxon test are.

In the analysis of principal components and Pearson's correlation analysis, the interpretation is scarce and the values of variance explained in each axis are lacking.

Regarding Pearson's correlation analysis, the p-values are missing.

Author Response

thank you. please see attachment

Author Response File: Author Response.pdf

Reviewer 2 Report

First, I would like to congratulate the authors for the good work they have done. The authors have spent a great deal of time and effort in acquiring a large amount of data and analysing it extensively. The article is relevant and aligns perfectly with the aims and scope of Water.

Regarding the use of English, I understand how difficult it is to write in English for non-native speakers. The vocabulary and terminology used in the article are generally correct, however, I think the text needs to be reviewed by a native English speaker. There are parts of the text that need serious linguistic retouching. If the expression is improved and the use of the general language is reviewed, the article would improve a lot, since its content is great.

In reference to specific text changes, I propose the following:

• Line 17: Remove space in “the me”.
• Line 26: “Allium cepa” should be abbreviated. Use cepa through the text after its first mention.
• Line 38: Change “Im” to “In”.
• Line 86: Change “mechanism” to “metabolism”.
• Line 104: "Copper" should not be capitalized. The same for the rest of chemical elements.
• Line 118: “A. cepatest”. Space.
• Table 1: The dot (.) must be used as a decimal separator in all tables. It is used in the text. Homogenize.
• Table 1: “levelofthewater”. Space.
• Table 2: “TCC: Thermo Tolerant Coliforms”. Change to “TTC: Thermotolerant Coliforms”.
• Line 286: Change “Escherichai coli” to “Escherichia coli”. Same for line 524.
• Line 303: “Enterobacter aerogenes” should be in italics.
• Line 305: “Enterobacter aerogenes” should be abbreviated.
• Figure 2: Change “Gram – negative” to “Gram-negative”.
• Table 3: Remove (%) from TN.
• Table 4: “Resultsoftheanalysisofmetalspresent in thewaterof”. Fix format.
• Line 399: Change “Allium Cepa” to A. cepa”.
• Line 452: Throughout the text the term "Figure" has been used with a capital letter. Homogenize.
• Line 571: The subject of the sentence is missing. Rephrase.
• Line 605 and 635: English expression needs to be improved.
• Line 666: “moderatelystrong”. Fix it.
• Line 698: Why Supplementary Materials are “Not applicable”? You have nice Supplementary Materials.
• Line 705: Change “by” to “buy”.
• Line 796 and 797: check format and add a full stop at the end.

Author Response

thank you. please see attachment

Author Response File: Author Response.pdf

Reviewer 3 Report

The manuscript deals with the physicochemical and microbiological studies of the waters of Lake Buritis in Brazil. The authors also investigated the potential cytotoxic and genotoxic effects of contaminants present in the collected water samples.

After reading the manuscript and the corrections made by the authors under the influence of previous reviews, I conclude that the work may be published with some minor changes:

1. The title of the work is very complicated. The authors wanted to include all their research in it. This should be included in the keywords. And so in the title, for example, it was stated that the mutagenic potential of water samples was tested, and yet Allium assay is used to assess the toxicity, cytotoxicity, genotoxicity and mutagenicity of individual chemical substances, as well as complex mixtures.
2. The introduction needs to be improved. This part is too long. The authors unnecessarily describe in detail the unfavorable effects of the tested parameters, such as the hardness and acidity of water or the tested metals. These messages can be moved to the discussion. In the introduction, please describe only the local problem against the background of the available literature and global problems with water purity.
3. It is necessary to distinguish between the results of tests based on standard procedures regulated by law, and the results of non-standard tests, eg Allium assay, identification of microorganisms or their antibiotic resistance. The discussion in this regard should be expanded. Do the authors, for example, based on the results of their own and other researchers, suggest introducing some research into legal provisions in water quality research? This should be included in the summary in relation to the purpose of the work.
4. Please clearly define what parameters of the tested water samples were determined on the basis of the mitotic index (MI) and aberrations index (CAI). The authors use the terms mutagenicity and genotoxicity interchangeably, as well as toxicity and genotoxicity.
5. 177 bacterial strains have been isolated and antibiotic resistance has been tested for 102 isolates - why? There is no legend for table S4.
6. Tables S3 and S4 are missing spaces between generic and species names in places.
7. In lines 688-690 the authors state: Considering the data on root growth and especially MI values, a cytotoxic potential for the spring and mouth water of Buritis Lake, in Goiatuba city, Goiás state, Brazil, is suggested. On the other hand, mutagenicity for the lake, sampled at the mentioned points, is not assumed. The studies were carried out with only one test, perhaps, for example, the mutagenic potential of the pollutants would be detected after the application of the Salmonella test. The results of these observations should be formulated differently.
8. The microbiological methods is only based on classical microbiology, which cannot identify the non-cultivated microorganisms. It is suggested to add some contents about the limitation of this study, and point out using high-throughput sequencing techniques for supplementation.

Author Response

thank you. please see attachment

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

• The following sentence remains incomprehensible:

“Principal Component Analysis (PCA)with centralization of the mediaand Pearson correlation was applied to verify the relationships of the two results obtained two physical-chemical parameters and test of A. cepa, thus finding groups that have similar behaviors or correlations, in this case, was used STATISTICA software version 10.0 (Statsoft, Tulsa, Ok, USA)”

• As far as statistical analysis is concerned, there is still some confusion, as in grammatical terms the sentences are not clear. The authors refer the test for data normality, but they do not mention anything about the homogeneity of variances, which is a crucial point for carrying out a parametric or non-parametric test. Thus, it is necessary to indicate the validation of the existence (or not) of the homogeneity of variances.

 

• On the other hand, they state that:

“For analysis of normality, the Shappiro-Wilk and Kolmogorov-Smirnov tests were performed according to the sample size; the normality tests were associated with the evaluation of measures of central tendency and graphical distributions. For acceptance of the significance of the normality tests, the significance limit of 5% was considered, thus, when the tests presented p-value ≤ 0.05, the data were considered non-parametric and, when they presented p-value > 0.05, they were considered parametric data. Subsequently, the Tukey test for parametric data and Games-Howell test for non-parametric data were performed, thereby determining the differences between means of three groups.”

However, the Tukey test and/or the Games-Howell test are only performed when the null hypothesis of the analysis of variance (ANOVA) or the Kruskal-Wallis test (depending on whether the data meet the requirements of the analysis of variance or not) is rejected. Nevertheless, the authors make no mention of the analysis of variance test (whether it was ANOVA or Kruskal-Wallis). This information must be clearly and correctly described.

• “For the analysis of the A. cepa test results, was used ANOVA” » The following sentence is still not correct, because what is analyzed with a statistical test are data and not results. On the other hand, to apply an analysis of variance it is necessary to validate both the normality and the homogeneity of variances. However, and once again the authors mention nothing about this. This information must be clearly described.

 

• There is still no consistency in the presentation of results, as can be seen from this paragraph. “The results of the March 2018 analyzes, showed that the root growth of the meristematic cells of A. cepa, showed a significant difference (p-value< 0.0001), between the samples and the negative control. The post hoc test (p-value<0.05), indicated the signifi- 420 cant difference of the P2 point in relation to the negative control. The MI indicated a sig- 421 nificant difference between the negative control and the P2 point (p-value = 0.008). The 422 post hoc analysis (Games - Howell test), of the treatment, showed a significant difference 423 for the sample of point P2 (p-value ≤ 0.05).><0.05), indicated the significant difference of the P2 point in relation to the negative control. The MI indicated a significant difference between the negative control and the P2 point (p-value = 0.008). The post hoc analysis (Games - Howell test), of the treatment, showed a significant difference for the sample of point P2 (p-value ≤ 0.05)” »» Why write p-value<0.05 and in another part of the sentence write p-value=0.008?

 

• Concerning my previous comment “Still in the presentation of the results, the authors sometimes identify the multiple comparison tests used, but other times they do not.”, the authors did not respond or act to clarify my comment. What has been answered (“Statistical tests were applied according to ANOVA. The multiple comparison is described at the of the results (topic 3.6)”), makes no sense. They must effectively respond to this comment.

Author Response

Reviewer 1

Report 2

 

“Principal Component Analysis (PCA)with centralization of the mediaand Pearson correlation was applied to verify the relationships of the two results obtained two physical-chemical parameters and test of A. cepa, thus finding groups that have similar behaviors or correlations, in this case, was used STATISTICA software version 10.0 (Statsoft, Tulsa, Ok, USA)”

As far as statistical analysis is concerned, there is still some confusion, as in grammatical terms the sentences are not clear. The authors refer the test for data normality, but they do not mention anything about the homogeneity of variances, which is a crucial point for carrying out a parametric or non-parametric test. Thus, it is necessary to indicate the validation of the existence (or not) of the homogeneity of variances.

Author´s answer: Shapiro-Wilk and Kolmogorov-Smirnov tests were performed to determine normality, to complement and increase assertiveness in choosing the comparison test, the Levene test was also performed to determine the significance of homogeneity. When the Shapiro-Wilk, Kolmogorov-Smirnov and Levene tests presented p ≤ 0.05, the data were considered non-parametric and non-homogeneous. The information was added in the methodology.

 

 On the other hand, they state that:

“For analysis of normality, the Shappiro-Wilk and Kolmogorov-Smirnov tests were performed according to the sample size; the normality tests were associated with the evaluation of measures of central tendency and graphical distributions. For acceptance of the significance of the normality tests, the significance limit of 5% was considered, thus, when the tests presented p-value ≤ 0.05, the data were considered non-parametric and, when they presented p-value > 0.05, they were considered parametric data. Subsequently, the Tukey test for parametric data and Games-Howell test for non-parametric data were performed, thereby determining the differences between means of three groups.”

However, the Tukey test and/or the Games-Howell test are only performed when the null hypothesis of the analysis of variance (ANOVA) or the Kruskal-Wallis test (depending on whether the data meet the requirements of the analysis of variance or not) is rejected. Nevertheless, the authors make no mention of the analysis of variance test (whether it was ANOVA or Kruskal-Wallis). This information must be clearly and correctly described.

Author´s answer: The variances of cytotoxicity detected by the A. cepa test between the collection points and the negative control were determined using the Analysis of Variance test (ANOVA) associated with the Games-Howell post hoc test when the data showed non-significant homogeneity and Tukey test when the data showed significant homogeneity, characterized by the Shapiro-Wilk and Levene tests.The information was added in the methodology.

 

“For the analysis of the A. cepa test results, was used ANOVA” » The following sentence is still not correct, because what is analyzed with a statistical test are data and not results. On the other hand, to apply an analysis of variance it is necessary to validate both the normality and the homogeneity of variances. However, and once again the authors mention nothing about this. This information must be clearly described.

Author´s answer: Analysis of Variance test (ANOVA) associated with the Games-Howell post hoc test when the data showed non-significant homogeneity and Tukey test when the data showed significant homogeneity, characterized by the Shapiro-Wilk and Levene tests.The information was added in the methodology.

 

There is still no consistency in the presentation of results, as can be seen from this paragraph. “The results of the March 2018 analyzes, showed that the root growth of the meristematic cells of A. cepa, showed a significant difference (p-value< 0.0001), between the samples and the negative control. The post hoc test (p-value<0.05), indicated the signifi- 420 cant difference of the P2 point in relation to the negative control. The MI indicated a sig- 421 nificant difference between the negative control and the P2 point (p-value = 0.008). The 422 post hoc analysis (Games - Howell test), of the treatment, showed a significant difference 423 for the sample of point P2 (p-value ≤ 0.05).><0.05), indicated the significant difference of the P2 point in relation to the negative control. The MI indicated a significant difference between the negative control and the P2 point (p-value = 0.008). The post hoc analysis (Games - Howell test), of the treatment, showed a significant difference for the sample of point P2 (p-value ≤ 0.05)” »» Why write p-value<0.05 and in another part of the sentence write p-value=0.008?

 Author´s answer: The description of results has been improved for better understanding.

 

Concerning my previous comment “Still in the presentation of the results, the authors sometimes identify the multiple comparison tests used, but other times they do not.”, the authors did not respond or act to clarify my comment. What has been answered (“Statistical tests were applied according to ANOVA. The multiple comparison is described at the of the results (topic 3.6)”), makes no sense. They must effectively respond to this comment.

Author´s answer: The description of results has been improved for better understanding.

Author Response File: Author Response.docx