Review Reports - Integrating LASSO and Extreme Gradient Boosting for Optimal Multiple Linear Regression Modeling of Milk Color Traits in Relation to Somatic Cell Count and Milk Composition in Dairy Cows

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

This research provides valuable information regarding the correlations between color, somatic cell count and different physicochemical milk parameters. The most valuable conclusion is that there is strong correlation of SCC with conductivity, and that its limited correlation with color parameters suggests that inflammation causes ionic shifts, not pigment changes. The study has some limitations; however, the authors have clearly acknowledged them and suggested directions for future research to address them. There are a few repetitive paragraphs in the text..

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

We would like to sincerely thank you for your careful evaluation and constructive comments on our manuscript. Your suggestions have significantly improved the clarity, scientific accuracy, and overall quality of the paper. We have addressed all comments point-by-point as detailed below.

Comment 1:
There are repetitive paragraphs in the text.

Response:
Thank you for this observation. Repetitive sentences and paragraphs (particularly in the Results and Discussion sections) have been carefully identified and removed. The manuscript has been revised to improve clarity and avoid redundancy. (Lines 232–235; 320–340)

Comment 2:
The color cannot affect the milk components; it is the other way around.

Response:
We appreciate this important clarification. The relevant sentence in the Introduction has been revised to correctly reflect the causal relationship. The text now clearly states that milk color reflects underlying physicochemical properties rather than influencing milk composition. (Lines 71–73).

Comment 3:
Use “content” instead of “concentration”.

Response:
Thank you for this suggestion. The terminology has been corrected throughout the manuscript. All instances of “concentration” referring to milk components have been replaced with “content” to align with standard dairy science terminology. (Lines 115–120)

Comment 4:
Use “physicochemical properties” instead of “components” where appropriate.

Response:
We agree with this comment. The terminology has been revised accordingly. The manuscript now distinguishes clearly between milk composition (e.g., protein, fat, lactose) and physicochemical properties (e.g., electrical conductivity, freezing point). (Lines 138–142)

Comment 5:
Minor grammatical and punctuation corrections.

Response:
All indicated issues have been corrected, including punctuation (e.g., use of colon instead of semicolon) and missing commas. The manuscript has also been carefully proofread to improve overall language quality. (Line 141 and Line 198)

Comment 6:
Figure 2 legends are not aligned with the images.

Response:
Thank you for pointing this out. Figure 2 has been revised to ensure correct alignment between figure panels and their corresponding labels (a–d). The figure caption has also been clarified. (Lines 210–220)

Comment 7:
Lines 252–254 should be transferred to the Discussion section.

Response:
We agree with this suggestion. The relevant explanatory sentences have been moved from the Results section to the Discussion section to ensure proper separation between results and interpretation. (original Lines 250–254)

Once again, we sincerely thank you for your valuable feedback, which has greatly improved our manuscript.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors, Respectfully, I believe this manuscript demonstrates potential. However, several issues currently limit its overall quality and clarity. The most significant concerns are the generic and overly broad writing style, the absence of clearly defined novelties, and the lack of a single, focused objective around which the study is developed. These issues reduce the scientific impact of the work and diminish its overall relevance.

Abstract:

Lines 30-33: I do not agree with the conclusion. Based on my observations, only chroma provides potentially relevant information; however, chroma is not a phenotypic trait, as it requires specific equipment and calculations for its estimation.

Keyword: My suggestion is to use keywords other than the title and review the keyword suggestion in the author's instructions.

Introduction: I like this introduction because it has relevant information. My suggestion is to add numbers to improve your description because in the current form it appears a generic description. For example, MxH animals endure a climate that is 4 °C warmer than purebred animals, which is indicative of adaptation and low production of oxidant molecules.

Lines 46–47: Place “a” and “b” in quotation marks.

Lines 48–49: Add a reference to support this statement.

Lines 59–65: This section presents an extensive, unreferenced passage intended to introduce the absence of studies in this area. Please include appropriate references or revise the text to improve support and conciseness.

Lines 66–67: I recommend avoiding statements such as “Our study is the first,” “Few studies exist,” or “Data are minimal.” Rather than strengthening the manuscript, such expressions may inadvertently suggest limited scientific novelty. They can be perceived as appeals for validation rather than demonstrations of the study’s intrinsic merit. This manuscript contains several strong and original aspects that could be emphasized more effectively. Therefore, consider removing these statements or rewriting them to highlight the specific contributions of the study. The same recommendation applies to similar phrasing throughout the manuscript.

Lines 67–68: There is a lack of connection between paragraphs. Please revise to improve coherence and flow.

Lines 71–73: According to the title, the objective is to assess milk color as an indicator of milk quality; however, this is not clearly stated here.

Lines 73–75: Similarly, the hypothesis is not aligned with the title or the stated objectives in the introduction.

Lines 77–80: Is this an additional objective? The manuscript should present a single, clearly defined objective. Please avoid extended passages that read more as commentary.

Lines 80–84: This appears to introduce yet another objective. The manuscript should focus on a single objective and avoid lengthy, comment-like text.

Results:

Lines 185–186: I suggest removing this type of description, as it is redundant and overlaps with the table title. The same applies to similar descriptions in other tables and figures.

Lines 230–231: Please verify whether the R² value for fat variation is sufficient to support the conclusion that fat influences the “L”, “b”, and hue parameters. Alternatively, could the relationship be inverse? This point requires clarification.

Lines 243–244: This is an interesting finding and should be explained in greater detail.

Line 255: The expression “dominant role” may be inappropriate, as other relevant parameters affecting milk quality were not assessed in this study. Please be more specific and limit the interpretation to the variables evaluated.

Discussion:

Lines 266–268: This section should be rewritten, as the statement is not sufficiently supported by the results of this study.

Lines 268–269: The term “these components” may be inappropriate, since other relevant parameters affecting milk quality were not assessed. Please be more specific and restrict the interpretation to the evaluated variables.

Line 273: Please explain how the colloidal structure influences the optical properties, beyond the effect of chemical composition.

Lines 277–283: Given the extensive body of literature on this topic, it is unclear why these aspects were not considered in the interpretation of the results. Please address this gap.

Lines 342–344: It is important to clarify that chroma and conductivity showed stronger contributions only among the variables evaluated in this study. If additional variables were included, other parameters might present stronger correlations.

Conclusion: The conclusion largely compiles results rather than directly addressing the study’s main objective. As currently presented, the title, objective, and conclusion are not clearly aligned, which may lead to confusion for readers. The authors should first define a clear primary objective, then adjust the title accordingly and rewrite the conclusion to directly reflect this objective.

This alignment would also help strengthen the discussion of the results and, importantly, highlight the key contributions of the study. In addition, the introduction does not clearly articulate the study’s potential novelties. Stating that no prior studies exist does not, by itself, establish novelty; instead, the manuscript should emphasize the specific contributions and relevance of the research.

Author Response

Dear Reviewer,

We would like to sincerely thank the reviewer for the careful evaluation of our manuscript and for the constructive and insightful comments. We greatly appreciate the time and effort invested in reviewing our work. The suggestions have significantly improved the clarity, focus, and scientific rigor of the manuscript. Below, we provide a detailed, point-by-point response to each comment. All revisions have been carefully implemented in the manuscript.

Abstract:

Response 1:

We thank the reviewer for this valuable comment. We agree that the original statement in the Abstract may have been too strong and potentially misleading. In response, we have revised the concluding sentence to provide a more cautious and accurate interpretation of the findings. Specifically, we removed the expression “optical phenotype” and clarified that Chroma was the most informative variable among those evaluated, while emphasizing that color-based measurements should be considered as complementary rather than definitive indicators of milk quality.

Keyword: My suggestion is to use keywords other than the title and review the keyword suggestion in the author's instructions.

Response 2:

We appreciate this helpful suggestion. The keywords have been revised to avoid redundancy with the title and to better reflect the methodological and scientific scope of the study. More specific and indexing-friendly terms have been selected in accordance with the journal guidelines

Introduction: I like this introduction because it has relevant information. My suggestion is to add numbers to improve your description because in the current form it appears a generic description. For example, MxH animals endure a climate that is 4 °C warmer than purebred animals, which is indicative of adaptation and low production of oxidant molecules.

Response 3:

The introductory paragraph has been revised to include quantitative information, providing a more precise and scientifically grounded description. Numerical ranges related to milk fat content and somatic cell count were added to improve clarity and strengthen the contextual background.

Lines 46–47: Place “a” and “b” in quotation marks.

Response 4:

Thank you for noting this detail. The notation has been corrected, and the CIELab parameters are now consistently presented using appropriate formatting (e.g., “a*” and “b*”).

Lines 48–49: Add a reference to support this statement.

Response 5:

Appropriate references have been added to support the statement regarding light scattering by casein micelles and fat globules, ensuring that the claim is properly grounded in the literature.

Response 6:

We thank the reviewer for highlighting this issue. The paragraph has been substantially revised to improve conciseness and scientific support. General descriptions of machine learning have been removed, and the section has been rewritten to directly focus on relevant applications in dairy science, supported by appropriate references.

Response 7:

We fully agree with this recommendation. Statements suggesting a lack of studies have been removed or rephrased. Instead, the revised text now highlights the specific context and contribution of the present study without relying on claims of novelty based solely on limited prior research.

Lines 67–68: There is a lack of connection between paragraphs. Please revise to improve coherence and flow.

Response 8:

The transition between paragraphs has been improved by explicitly linking machine learning applications to the detection of subclinical mastitis. This revision enhances coherence and ensures a smoother logical flow.

Lines 71–73: According to the title, the objective is to assess milk color as an indicator of milk quality; however, this is not clearly stated here.

Response 9:

We appreciate this important observation. The objective will be completely relisted to clearly indicate the purpose of evaluating milk color as a potential optical variation of milk color. In line with our reviewer's opinion, the revised objective is now consistent with the overall scope of its title and content.

Lines 73–75: Similarly, the hypothesis is not aligned with the title or the stated objectives in the introduction.

Response 10:

The hypothesis has been revised to directly correspond with the stated objective. It now explicitly focuses on the relationship between milk color parameters and milk quality traits, ensuring conceptual consistency throughout the manuscript.

Lines 77–80: Is this an additional objective? The manuscript should present a single, clearly defined objective. Please avoid extended passages that read more as commentary.

Response 11:

We agree with the reviewer. The identified sentences were removed as they introduced unnecessary commentary and could be interpreted as additional objectives. The manuscript now presents a single, clearly defined objective.

Lines 80–84: This appears to introduce yet another objective. The manuscript should focus on a single objective and avoid lengthy, comment-like text.

Response 12:

The section has been removed to avoid redundancy and to maintain a clear and focused structure. The remaining text has been adjusted to align strictly with the main objective of the study.

Results:

Lines 185–186: I suggest removing this type of description, as it is redundant and overlaps with the table title. The same applies to similar descriptions in other tables and figures.

Response 13:

Redundant sentences describing the content of tables have been removed. The Results section now focuses on interpretation rather than repeating table titles.

Response 14:

Thank you for this important comment. The interpretation of fat-related results has been revised. We now explicitly acknowledge the relatively low explanatory power (R² = 0.191) and have moderated the conclusions accordingly, emphasizing the limited and indirect nature of the relationship.

Lines 243–244: This is an interesting finding and should be explained in greater detail.

Response 15:

The relevant section has been expanded to provide a more detailed explanation of the underlying mechanism, particularly the role of ionic changes (Na⁺, Cl⁻), refractive index variations, and their influence on light scattering and optical measurements.

Response 16:

The expression “dominant role” has been replaced with a more precise and cautious wording. The interpretation is now explicitly limited to the variables evaluated within this study.

Discussion:

Lines 266–268: This section should be rewritten, as the statement is not sufficiently supported by the results of this study.

Response 17:

We appreciate this comment. The statement has been rewritten to avoid overinterpretation. The revised text now frames the findings as being consistent with existing knowledge rather than presenting them as direct evidence of underlying mechanisms.

Response 18:

The wording has been revised to explicitly refer to “protein and fat variables evaluated in this study,” thereby avoiding overly general statements.

Line 273: Please explain how the colloidal structure influences the optical properties, beyond the effect of chemical composition.

Response 19:

We thank the reviewer for this valuable suggestion. The discussion has been expanded to include a mechanistic explanation of how colloidal structure influences optical properties, including particle size, refractive index differences, and light scattering behavior.

Lines 277–283: Given the extensive body of literature on this topic, it is unclear why these aspects were not considered in the interpretation of the results. Please address this gap.

Response 20:

The discussion has been strengthened by incorporating relevant literature and explicitly linking our findings to previous studies. This improves the scientific depth and contextual interpretation of the results.

Response 21 :

Thank you for this suggestion. The text has been revised to clearly state that the observed importance of Chroma and electrical conductivity is limited to the variables evaluated in this study, and that inclusion of additional parameters may alter these relationships.

Conclusion: The conclusion largely compiles results rather than directly addressing the study’s main objective. As currently presented, the title, objective, and conclusion are not clearly aligned, which may lead to confusion for readers. The authors should first define a clear primary objective, then adjust the title accordingly and rewrite the conclusion to directly reflect this objective.

Response 22 :

We sincerely thank the reviewer for this important comment. The Conclusion section has been completely rewritten to ensure full alignment with the study objective and title. The revised conclusion now directly addresses the role of milk color as an indirect optical indicator, avoids overgeneralization, and clearly states the limitations and practical implications of the findings.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

This paper investigates whether objective milk color measurements can predict key milk composition traits and help classify udder health status in Holstein cows. Using a single-herd dataset (n=119), the authors fit LASSO regressions to predict composition and an XGBoost classifier for SCC status, identifying Chroma and electrical conductivity as influential predictors.

The conceptual framing is sound and practically appealing, and the initial empirical results—while modest—suggest complementary value of colour (especially Chroma) with conductivity for mastitis-related screening. However, several methodological and reporting issues currently limit the paper’s reliability and impact.

Please see additional comments/suggestions in the attached pdf.

Comments for author File: Comments.pdf

Author Response

Dear Reviewer,

We sincerely thank Reviewer for the thorough, insightful, and constructive evaluation of our manuscript. The reviewer’s comments have significantly improved the clarity, methodological rigor, and overall quality of the study. We have carefully addressed each point and revised the manuscript accordingly. Below, we provide a detailed, point-by-point response.

Comment 1:

The sampling description is insufficient.

Response 1:

We thank the reviewer for this valuable comment. The sampling procedure has been clarified in Section 2.2. Specifically, we explicitly stated that milk samples were collected once per cow during the evening milking, and that each cow contributed a single observation. This clarification ensures transparency regarding sampling design and sample size.

(Lines 120–135)

Comment 2 :

Risk of preprocessing-related data leakage is unclear.

Response 2:

We appreciate this important observation. To address the concern, the preprocessing workflow has been revised in Section 2.3. We explicitly stated that all preprocessing steps (including transformation parameter estimation and standardization) were performed using only the training data and subsequently applied to validation/test sets. This ensures that no information leakage occurred during model development.

(Lines 150–190)

Comment 3:

Unclear whether repeated measurements were present.

Response 3:

The manuscript has been clarified to explicitly state that each cow contributed only a single observation and that no repeated measures structure was present. Accordingly, all observations were treated as independent in the analysis.

(Lines 150–152)

Comment 4:

Hue definition/calculation requires clarification.

Response 4:

We appreciate this comment. The description of hue has been revised in Section 2.2. The definition was simplified and clarified to avoid ambiguity, and an additional statement was included in the Discussion noting that hue values may vary depending on the computational approach used for angular transformation. This ensures consistency and avoids potential misinterpretation.

(Lines 135–140)

Comment 5:

Clarify SCC threshold and classification.

Response 5:

The SCC classification procedure has been clarified below Table 4. We explicitly stated that a threshold of 200,000 cells/mL was used to define subclinical mastitis, which is widely accepted in the literature. The classification of samples into low and high SCC groups has been clearly described.

(Lines 300–310)

Comment 6:

Performance metrics require clarification and consistency.

Response 6:

We appreciate this important comment. The performance metrics reported in the manuscript have been carefully revised to ensure consistency and clarity. Redundant and potentially confusing metrics (e.g., recall/precision inconsistencies) were removed, and the reporting was standardized using accuracy, sensitivity, and AUC. Additionally, interpretative statements were revised to avoid overinterpretation and to reflect moderate predictive performance more appropriately.

(Lines 360–390)

Comment 7:

Results appear overinterpreted.

Response 7:

We thank the reviewer for this valuable insight. The Discussion section has been carefully revised to adopt a more cautious and balanced interpretation. Statements suggesting strong predictive ability were moderated, and emphasis was placed on the indirect and limited explanatory nature of color-based optical measurements. Limitations such as sample size and class imbalance were explicitly acknowledged.

(Lines 330–360)

Comment 8:

Clarify software used.

Response 8:

Thank you for your feedback. A typo occurred in the Methods section (Section 2.3), and it has been revised to clarify that all analyses, preliminary processing, and the forecasting section were performed using custom scripts written in Python. This clarification ensures continuity and transparency in the analysis's performance.

(Lines 148–150)

Comment 9:

high value in milk fat content

Response 9:

We thank the reviewer for this insightful comment. We agree that the reported values may appear relatively high for Holstein cows. However, several factors may explain this observation. First, animals with clinical were not intentionally included in the study population, as the focus was on apparently healthy cows. Second, milk samples were collected during the evening milking, when milk yield is typically lower and total solids (including fat content) may be relatively higher compared to morning milking. Additionally, the herd included cows at different stages of lactation, including individuals close to the dry-off period. It is well established that milk composition, particularly fat and dry matter, can increase during late lactation stages. These combined factors likely contributed to the relatively elevated values observed in this study. To improve clarity, a brief explanation has been added to the manuscript.

Once again, we sincerely thank Reviewer for the careful evaluation and constructive suggestions. The comments have greatly contributed to improving the scientific quality, clarity, and robustness of the manuscript. We believe that the revised version addresses all concerns and is now significantly strengthened.

Author Response File: Author Response.pdf

Round 2

Reviewer 1 Report

Comments and Suggestions for Authors

The authors addressed all issues, and I'm happy with their response. I recomend accept manuscript.

Author Response

Comment :

The authors addressed all issues, and I'm happy with their response. I recomend accept manuscript.

Response :

We sincerely thank the reviewer for their invaluable contributions and suggestions that helped our article reach this level. Their feedback significantly improved the simplicity and clarity of our article.

Reviewer 2 Report

Comments and Suggestions for Authors

I am agree with the changes made it by the authors

Author Response

Comment:
I am agree with the changes made by the authors.

Response:
We sincerely thank the Reviewer for the positive evaluation of our revised manuscript. We appreciate the reviewer’s acknowledgment of the changes made and are pleased that the revisions have met their expectations.

Reviewer 3 Report

Comments and Suggestions for Authors

I thank the authors for considering my suggestions.
However, some minor issues remain (see pdf).

Comments for author File: Comments.pdf

Author Response

Comment:
I thank the authors for considering my suggestions. However, some minor issues remain (see pdf).

Response:

We sincerely thank our reviewer for carefully considering our article and taking our corrections into account. We appreciate the reviewer's constructive feedback and have addressed all minor issues noted in the annotated PDF. The labeling issue related to "L*" in Figure 3 has been corrected to ensure clarity and consistency throughout the article.