Review Reports - High-Definition Brain Network (HDBN) Delineation of CDKL5 Deficiency Disorder (CDD) in Genetically Engineered Mice

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

In the paper entitled “High-definition Brain Network (HDBN) Delineation of CDKL5 Deficiency Disorder (CDD) in Genetically Engineered Mice”, West et al. report that they established high-definition brain network (HDBN) biomarkers for advanced diffusion MRI tractography to quantify whole-brain network organisation in Cdkl5 knockout mice. The aim of the study was to establish clinically relevant brain-based biomarkers for therapeutic testing in Cdkl5 KO mice, large animals, and humans.

The study is generally well designed. However, some aspects of the analysis, interpretation, and rewriting require attention.

Major points:

The Funding section is missing.
Lines 232–236 should not be underlined.
Explain the colour coding in Figure 1B.
Figures 1F to 1H: Enlarge the font and provide higher resolution images so that inscriptions are readable.
Line 278: There is no need to refer to the figure here.
Figure 2 legend: Define MD, AD, RD, etc. here as well.
Figure 3: Provide a higher resolution image, as nothing can be deduced from these panels. Perhaps you submitted a higher resolution image and the poor quality is due to PDF conversion during submission. The same comment applies to all Figures from 4 to 13.
Figures are poorly referenced in the text.
Chapter 3.4.2, “ Regional network topology changes in Cdkl5 brains”, is difficult to read. Describe the data more clearly.
There are no p-values in the figure legends. Also, the number of animals used in each experiment is not mentioned in the figure legends.
Figure titles should be informative by describing the results that can be deduced from the data shown in the panels.
There is no text between most images, making the manuscript difficult to read.
There are too many references. Retain only the most relevant and recent ones.
Define “imaging biomarker” more clearly at the beginning.

Author Response

We would like to thank the reviewer for their thoughtful and very constructive feedback. We appreciate the time and effort invested in evaluating our manuscript. These insightful comments have helped us improve the manuscript greatly. we have carefully considered all suggestions and have revised the manuscript accordingly. we believe these changes have greatly strengthened the overall quality of the study and are deeply grateful for the reviewer's insight. Below are the comments and our rebuttals to them referencing the in-manuscript line numbers. Changes in the manuscript are tracked in red.

The Funding section is missing.
1. We added a funding section at lines 1140-1142
Lines 232–236 should not be underlined.
1. Thank you for pointing this out; we have removed the underlined words in lines 232-236.
Explain the colour coding in Figure 1B.
1. Added color coding description for 1B in figure legend located at lines 276-277.
Figures 1F to 1H: Enlarge the font and provide higher resolution images so that inscriptions are readable.
1. Enlarged figure resolution to 900DPI and made a new figure to increase the font size.
Line 278: There is no need to refer to the figure here.
1. Thank you, the reference to Figure 1 has been removed.
Figure 2 legend: Define MD, AD, RD, etc. here as well.
1. In Figure 2 legend, we added the definitions of FA, MD, AD, and RD.
Figure 3: Provide a higher resolution image, as nothing can be deduced from these panels. Perhaps you submitted a higher resolution image, and the poor quality is due to PDF conversion during submission. The same comment applies to all Figures from 4 to 13.
1. Thank you, we created new figures to allow for an increase in the size of text. We also increased the image quality to 900dpi.
Figures are poorly referenced in the text.
1. Referenced figures in text.
Chapter 3.4.2, “ Regional network topology changes in Cdkl5 brains”, is difficult to read. Describe the data more clearly.
1. Rewrote chapter 3.4.2 so that each region was talked about prior to their respective figure. This also added more text between figures to help with difficulty reading.
There are no p-values in the figure legends. Also, the number of animals used in each experiment is not mentioned in the figure legends.
1. Referenced supplemental tables for the P-values within the figure legends. Added animal numbers to figure legends as well.
Figure titles should be informative by describing the results that can be deduced from the data shown in the panels.
1. Thank you for bringing this up; we have added figure titles describing the results to each figure.
There is no text between most images, making the manuscript difficult to read.
1. Thank you for pointing this out; we have added text about the referenced regions prior to the figure.
There are too many references. Retain only the most relevant and recent ones.
1. Thank you for pointing this out, we decreased the reference amount and adjusted them so that they match the journal's recommendations.
Define “imaging biomarker” more clearly at the beginning.
1. Defined “imaging biomarker” in the introduction on line 73-74.

Reviewer 2 Report

Comments and Suggestions for Authors

Dear authors,

Your manuscript is original and makes valuable contributions to the field of treatment-resistant epilepsy. However, as a reviewer, I would like to offer some comments on a few key points in each section, which I hope will be helpful.

The abstract is well-structured and clearly communicates the translational relevance of the study. I recommend considering the following:

Some sentences are lengthy and could be simplified to improve clarity.
Minor conceptual repetitions should be reduced.
The main finding could be more clearly emphasized in quantitative or differential terms.

The introduction provides an adequate conceptual framework and correctly justifies the need for brain biomarkers in CDD.

Suggestions:

Reduce some lengthy theoretical explanations (e.g., in graph theory) to maintain focus on the study hypothesis.
Review minor redundancies in the clinical description of CDD.
Correct some minor details of style and grammatical agreement.

Given that the study proposes HDBNs as translational biomarkers, could you elaborate on how these network parameters would differ from other structural biomarkers already used in neurodevelopmental disorders and what specific advantage they offer in the clinical context of CDD?

The methodology is described in sufficient technical detail and allows for reproducibility. In this regard, I recommend the following:

Review the consistency of the unit format (spaces between numbers and units).

Improve the wording of some technical phrases for greater clarity.
Briefly clarify the rationale for certain tractography parameters (e.g., thresholds and exclusion criteria) to strengthen methodological robustness.

Considering that the brains were analyzed ex vivo, how do you think PFA fixation and acquisition conditions might have influenced diffusion metrics and network quantification? How comparable would these results be with in vivo studies?

The results are extensive and technically sound. However:

This section could benefit from greater synthesis to avoid repetition of similar interpretations across different regions.
Some lengthy theoretical explanations could be moved to the Introduction or Discussion.
It is recommended to standardize terminology (e.g., “diffusion parameters,” “diffusion metrics,” “quantitative diffusivity”). The findings are consistent and biologically plausible, but more concise wording would improve clarity.

The study makes multiple regional and connectivity comparisons. In addition to adjusting for FDR, did you assess the stability of the findings using sensitivity analyses (e.g., by varying tractography or graph extraction thresholds)? This would help confirm the robustness of the results.

Some regions showed alterations in network metrics without global changes at the whole-brain level. How do you interpret this preservation of global organization in the face of significant regional alterations? Does this suggest a compensatory mechanism or selective circuit reorganization?

Given that murine models do not exhibit overt epileptic seizures, how do you propose to validate that changes in HDBN reflect progression or therapeutic response in longitudinal studies? What steps would be necessary to translate this biomarker into human clinical trials?

The figures are informative and technically well-prepared. I recommend:

Verifying consistency in legends (capitalization, repeated terms).
Simplifying some lengthy legends.
Ensuring that all abbreviations are defined in their corresponding figures or tables.

Overall, the graphic material is appropriate and adequately supports the results.

I suggest that the discussion and conclusions be considered separate sections. The discussion could be aligned with the order of the results to avoid repetition of certain concepts and could also highlight the strengths and weaknesses of the study at the end. I suggest differentiating between confirmatory and exploratory findings in the discussion.

The conclusions are supported by the data presented. I recommend:

Making them slightly more concise.
Avoiding overly broad statements about clinical applicability without clarifying limitations.

Finally, the reference list does not fully comply with the editorial format required by Biomolecules (MDPI). It was observed that several journal names are not abbreviated according to PubMed standards, the DOI format is inconsistent (some entries include “doi:” while others present the full URL), and title capitalization is not uniform. In general, the references follow a PubMed-style format more closely than the final format required by MDPI.

It is recommended to carefully review the entire references section to ensure its full compliance with the journal's editorial guidelines, including official journal abbreviations, consistent DOI presentation, and correct title capitalization.

Comments on the Quality of English Language

The manuscript contains minor grammatical errors, such as excessively long sentences, which I suggest be simplified to improve clarity and fluency. Terminological repetition and inconsistencies in the use of some technical terms are also noted. Moderate linguistic revision is recommended to optimize the accuracy and academic style of the text.

Author Response

We are grateful to the reviewer for their careful assessment of our manuscript, as well as the constructive criticism provided. The Comments were very insightful and have greatly contributed to improving the manuscript. We have addressed each point thoroughly and have revised the manuscript accordingly. We greatly appreciate the reviewers engagement with our work and firmly believe that the revisions will have enhanced the content and clarity of the manuscript. Below are the comments and our rebuttals to them referencing the in-manuscript line numbers. Changes in the manuscript are tracked in red.

Reduce some lengthy theoretical explanations (e.g., in graph theory) to maintain focus on the study hypothesis.
1. We rewrote the graph theory paragraph so that it wasn’t as lengthy.
Review minor redundancies in the clinical description of CDD.
1. Removed redundant sentences for the clinical description of CDD.
Correct some minor details of style and grammatical agreement.
1. Thank you, we have revised the manuscript for styling and grammatical agreements.
Given that the study proposes HDBNs as translational biomarkers, could you elaborate on how these network parameters would differ from other structural biomarkers already used in neurodevelopmental disorders and what specific advantage they offer in the clinical context of CDD?
1. We added in the current biomarkers and included that there was a lack of Brain based biomarkers on lines: 73-74
Review the consistency of the unit format (spaces between numbers and units).
1. Thank you, we adjusted all of the unit formats to make sure that they all matched.
Improve the wording of some technical phrases for greater clarity.
1. Improved word clarity to make technical phrases easier to follow.
Briefly clarify the rationale for certain tractography parameters (e.g., thresholds and exclusion criteria) to strengthen methodological robustness.
1. We added that the threshold was recommended via DSI studios on line 217.
Considering that the brains were analyzed ex vivo, how do you think PFA fixation and acquisition conditions might have influenced diffusion metrics and network quantification? How comparable would these results be with in vivo studies?
1. Thank you for pointing this out, we discussed an advantage of ex vivo imaging relative to this study on lines: 164-166.
This section could benefit from greater synthesis to avoid repetition of similar interpretations across different regions.
1. Thank you for this insight, we cut down on the amount of repeated interpretation across regions.
Some lengthy theoretical explanations could be moved to the Introduction or Discussion.
1. Thank you, theoretical explanations were moved to the discussion.
It is recommended to standardize terminology (e.g., “diffusion parameters,” “diffusion metrics,” “quantitative diffusivity”). The findings are consistent and biologically plausible, but more concise wording would improve clarity.
1. Thank you for pointing out this, we standardized termonology so that the wording was more concise.
The study makes multiple regional and connectivity comparisons. In addition to adjusting for FDR, did you assess the stability of the findings using sensitivity analyses (e.g., by varying tractography or graph extraction thresholds)? This would help confirm the robustness of the results.
1. Thank you for suggesting this, we state that we set a graph extraction threshold of 0.001 on line: 245
Some regions showed alterations in network metrics without global changes at the whole-brain level. How do you interpret this preservation of global organization in the face of significant regional alterations? Does this suggest a compensatory mechanism or selective circuit reorganization?
1. Thank you, this is a great point. We add to the discussion that there may be a compensatory process involving alternate pathways in the discussion section at lines: 974-977
Given that murine models do not exhibit overt epileptic seizures, how do you propose to validate that changes in HDBN reflect progression or therapeutic response in longitudinal studies? What steps would be necessary to translate this biomarker into human clinical trials?
1. Thank you, we provided next analysis steps of using our HDBN on genetically altered pigs to both validate our findings and make it more applicable to humans in the conclusion, lines: 1082-1083.
Verifying consistency in legends (capitalization, repeated terms).
1. Thank you for realizing there was some inconsistency. We updated the legends accordingly.
Simplifying some lengthy legends.
1. Thank you, we adjusted some figure legend lengths so that they were shorter.
Ensuring that all abbreviations are defined in their corresponding figures or tables.
1. Thank you for bringing this up, abbreviations are defined in a supplemental table #1.
I suggest that the discussion and conclusions be considered separate sections. The discussion could be aligned with the order of the results to avoid repetition of certain concepts and could also highlight the strengths and weaknesses of the study at the end. I suggest differentiating between confirmatory and exploratory findings in the discussion.
1. Thank you for this suggestion. We have created a conclusion section. We also adjusted how the discussion read so that it aligned more with the results in terms of order.
Making them slightly more concise.
1. Thank you for suggesting this, we trimmed down the conclusions to make it more concise.
Avoiding overly broad statements about clinical applicability without clarifying limitations.
1. Thank you for bringing this up, we reeled back the some of the clinical applications to make it less broad.
Finally, the reference list does not fully comply with the editorial format required by Biomolecules (MDPI). It was observed that several journal names are not abbreviated according to PubMed standards, the DOI format is inconsistent (some entries include “doi:” while others present the full URL), and title capitalization is not uniform. In general, the references follow a PubMed-style format more closely than the final format required by MDPI. It is recommended to carefully review the entire references section to ensure its full compliance with the journal's editorial guidelines, including official journal abbreviations, consistent DOI presentation, and correct title capitalization.
1. Thank you for noticing this. We have updated the references so that they are more inline with the journals guidelines.
The manuscript contains minor grammatical errors, such as excessively long sentences, which I suggest be simplified to improve clarity and fluency. Terminological repetition and inconsistencies in the use of some technical terms are also noted. Moderate linguistic revision is recommended to optimize the accuracy and academic style of the text.
1. Thank you, we have looked at the grammatical contents of the manuscript and corrected them.