All reviews of published articles are made public. This includes manuscript files, peer review comments, author rebuttals and revised materials. Note: This was optional for articles submitted before 13 February 2023.
Peer reviewers are encouraged (but not required) to provide their names to the authors when submitting their peer review. If they agree to provide their name, then their personal profile page will reflect a public acknowledgment that they performed a review (even if the article is rejected). If the article is accepted, then reviewers who provided their name will be associated with the article itself.
Many thanks for the revised manuscript. All four of our reviewers have confirmed that their comments have been addressed and showed their support for this publication.
Meanwhile, during this second round of review, some very minor grammatical/formality issues were noticed (detailed below). Hopefully, they can be of some help to you and our Production team before publishing this research paper:
1. If appropriate, kindly change “gene” in lines 31 and 39 to “genes.”
2. Kindly insert the appropriate formula to line 97.
3. If appropriate, kindly start a separate paragraph for the sentences in lines 310-315.
4. There is a duplicate word “overall survival” in line 497.
5. Kindly verify the capital/non-capital font for the titles of Figures 4-6 and 8.
[# PeerJ Staff Note - this decision was reviewed and approved by Paula Soares, a PeerJ Section Editor covering this Section #]
The authors has addressed my questions. No further comments.
The authors has addressed my questions. No further comments.
The authors has addressed my questions. No further comments.
The authors has addressed my questions. No further comments.
The manuscript by Hou, Chen et al. meets the basic reporting standards. The study on chemokine-related biomarkers in AML is clearly written, with appropriate terminology and a thorough literature review. The structure is professional, with relevant figures and tables. The authors have provided the raw data, addressing previous comments.
The research is original, well-defined, and relevant, falling within the aims and scope of PeerJ. The investigation is rigorous, with methods described in sufficient detail for replication. The initial selection of CCRG genes has been justified, although a clearer rationale could be provided. Overall, the study design and execution are robust.
The findings are robust and statistically sound. The conclusions are well-supported and linked to the research question. The authors have addressed previous comments regarding figure presentation and biomarker expression discussions. The addition of a supplementary figure enhances reader comprehension.
The authors' revisions have satisfactorily addressed my previous comments. The manuscript is now suitable for publication in PeerJ. I approve the publication of this manuscript.
In their rebuttal letter, the authors explained that they had limited time to perform additional experiments requested by reviewers, which is understandable. The conclusion has been tuned down and I find the revised manuscript acceptable for publication.
NA
NA
NA
The authors improved the writing of this manuscript a lot. The results and conclusions became more clear. But there are still a few mistakes. In line97, there is one more period. In line 161, analysis should be used instead of analyses. So it's better to further proof read the manuscript.
My comments have been addressed. The selection of different analysis became more clear.
In general, my comments have been addressed. They added some discussion about the potential functions of the DE-CCRGs. In addition, the authors need to be careful about the conclusions as many of them require further experimental validation.
For the figures, it seems like the authors randomly chose different fonts for the labels and the titles. I suggest to use consistent font and format for all the text in all the figures. The authors also need to proofread the figures. In figure 2E, the space between high risk and low risk is missing.
To identify prognostic chemokines involved in AML, the authors performed a bioinformatic analysis using three publicly available data sets and a qPCR assay using patients' samples, thus potentially benefiting future studies and patients. Yet, our reviewers have raised some comments, and the authors’ responses would be really appreciated.
Additionally, and as hinted by some reviewer(s), I would appreciate some amendments clarifying if the finding shows correlation vs. causation, such as lines 36 and 179-180. Also, kindly provide the full names of the acronyms upon their first appearance and continue to use them from there. For example, AML in lines 47 and 51, the full names of DEGs, DE-CCRGs, GO, KEGG, LASSO, ROC, and GSVA identified in figure legends. Further, kindly verify the uses of capital/non-capital letters for consistency, the word “and,” space before some brackets, and some other minor grammatical matters. See, for example, in the title, as well as in lines 49, 54, 59, 61, 65, 80, 84, 86, 88, 95, 110, 119, 120, 122, 123, 219, 228, 230, 233, 235, 237, 240, 242, 248, 255, 258, 266, 269, 277, 279, 284, 292, 295, 297, 321, and Figure 3F as identified in the figure legend.
[# PeerJ Staff Note: The review process has identified that the English language should be improved. PeerJ can provide language editing services - please contact us at [email protected] for pricing (be sure to provide your manuscript number and title) #]
The author claimed that "In this study, we first focused on the expression levels and prognostic roles of CCRGs in AML (Line 223)", which is actually the content of their last figure (Figure 8). Discussion part should be written in alignment with the results section.
no comment
The reviewer appreciates the comprehensive data analysis performed to establish the prognostic risk model for AML and identify biomarkers. However, the author should provide more detailed information about the statistics and reproducibility of the data. Specifically, the author should include the subsection "Statistics and reproducibility" in Method section, and claim that all data represent the mean ± SEM or mean ± SD from ≥3 independent experiments (independent biological replicas) for each condition. The D’Agostino-Pearson omnibus normality test should be used to determine whether data are normally distributed. Datasets with gaussian distributions should be compared using Student’s t-test (two-tailed) or one-way ANOVA followed by Tukey’s post hoc test. For comparing non-Gaussian distributions, the nonparametric Mann–Whitney U test or Kruskal–Wallis (with post hoc Dunn) should be used for comparisons between two or more groups, respectively. The exact p-values and how stars were defined should be included in all figure legends.
In this manuscript, Hou et al. performed various analyses to identify biomarkers that might play key roles in AML prognosis and the tumor immune microenvironment. The major finding of this manuscript is that a prognostic risk model for AML was successfully established. Moreover, the authors identified six biomarkers (CXCR3, CXCR2, CXCR6, CCL20, CCL4, and CCR2) and incorporated in the risk model, providing insights into the treatment of AML. However, additional data and analysis are required to support and validate their findings.
Major Concerns:
1. The reviewer appreciates the comprehensive data analysis performed to establish the prognostic risk model for AML and identify biomarkers. However, the author should provide more detailed information about the statistics and reproducibility of the data. Specifically, the author should include the subsection "Statistics and reproducibility" in Method section, and claim that all data represent the mean ± SEM or mean ± SD from ≥3 independent experiments (independent biological replicas) for each condition. The D’Agostino-Pearson omnibus normality test should be used to determine whether data are normally distributed. Datasets with gaussian distributions should be compared using Student’s t-test (two-tailed) or one-way ANOVA followed by Tukey’s post hoc test. For comparing non-Gaussian distributions, the nonparametric Mann–Whitney U test or Kruskal–Wallis (with post hoc Dunn) should be used for comparisons between two or more groups, respectively. The exact p-values and how stars were defined should be included in all figure legends.
a. For quantitative real-time PCR results in Figure 8B, the author should plot individual data point, and the experiment should be more than 3 independent biological replicas for each condition to draw any conclusion.
b. Also, no significant difference between control an AML group regarding CXCR2 level (p=0.4108), which is not aligned with line 212.
2. From KM plot results (Figure 4), the author concluded that the high expression of CCL4, CCR2, CXCR2, and CXCR3 reduced the overall survival (OS) of the high-risk group, while the high expression of CCL20 and CXCR6 increased the OS (Line 178 to 180). While Figure 8 shows that all six biomarkers expression either significantly downregulated or show the trend of downregulation in AML compared to control group. The author should explain more in detail in the Discussion section why high expression of CCL4, CCR2, CXCR2, and CXCR3 reduced OS of the high-risk group, however higher expressions of CCL4, CCR2, CXCR2, and CXCR3 were observed in control group compared to AML group.
3. Sample size of each group should be similar to enable a fair comparison and generate valid statistical analysis. The author mentioned that "Transcriptome data of 194 AML and 20 control bone marrow mononuclear cell samples measured using the GPL17586 platform from the GSE114868 dataset were used for differential gene analysis and expression validation" in Line 76 to 78. However, sample size of each group should be similar to enable a fair comparison and generate valid statistical analysis. The significance observed in Figure 8A could be due to the sample size discrepancy.
4. The author claimed that "In this study, we first focused on the expression levels and prognostic roles of CCRGs in AML (Line 223)", which is actually the content of their last figure (Figure 8). Discussion part should be written in alignment with the results section.
Minor Concerns:
1. The author should include sample size data of each group in Figure 4, and also the unit of Time (months or years) in the plot.
2. The author should specify how the statistical significance of differences were calculated for Figure 7B and E, and include p value data in figure legend. In addition, color code of high and low risk group should align with other figure: high should be red and low should be blue, for better visualization purpose.
The author should correct the typo of "Group" in Figure 8A. The author should define the four stars in figure legend, and specify how the statistical significance of differences was calculated. To align with other figures in the manuscript, data of healthy group, served as control, should be plotted on the left side, for better visualization purpose.
This study by Hou, Chen et al focuses on the identification and prognostic analysis of chemokine-related biomarkers in Acute Myeloid Leukemia (AML). The authors have carried out differential expression analysis and functional enrichment analysis to pinpoint chemokine-related genes that are differentially expressed in AML. Subsequent Cox regression analysis helped in identifying significant biomarkers and in constructing a prognostic model. The research highlights CXCR3, CXCR2, CXCR6, CCL20, CCL4, and CCR2 as key biomarkers integral to AML prognosis and the tumor immune microenvironment, offering new insights into AML's prevention and treatment.
Clarity and Language:
The manuscript is clearly written and uses unambiguous language, with technical terminology appropriate for its scientific field.
Literature and Context:
The paper adequately reviews pertinent literature, providing sufficient background to position the study within the existing AML research framework. References are properly cited to substantiate the claims presented.
Structural Coherence:
The manuscript meets professional structural standards. The figures and tables are relevant, clear, and well-described. However, the provision of raw data as per journal policy would enhance the paper's value.
Self-Containance:
The paper is self-contained, presenting a coherent narrative that covers all results and findings relevant to the tested hypotheses. The subdivision of the content is appropriate.
Originality and Scope:
The research question is original, clearly articulated, relevant, and aligns with the aims and scope of PeerJ. It addresses a distinct knowledge gap regarding the role of chemokine-related biomarkers in AML. The analyses build upon existing microarray datasets and clinical information, enriching the field's understanding as presented to PeerJ's readership. A concern is the initial selection of CCRG genes, which requires clearer justification to avoid the Texas sharpshooter fallacy. The selection of only 29 out of 6743 DEGs, which were not prominent in the initial gene ontology analysis (as noted in Figure 1B, where the points should be smaller due to overlap), calls for a stronger rationale either through more detailed introductions to the function of these 29 genes in AML or additional bioinformatics support.
Rigor and Ethics:
The experimental procedures and ethical considerations adhere to high technical standards. The methods are sufficiently detailed to allow for replication.
Robustness of Data:
The data are robust, statistically sound, and derived from well-controlled, previously published datasets.
Strength of Conclusions:
The conclusions are strongly supported by the results and are directly linked to the research question. The assertions of causative relationships are well-founded based on the conducted analyses.
Figures 3A, 2D, and 6B would benefit from patient clustering within each group for clearer heatmap presentations, currently appearing random.
For Figure 8B, with fewer than 20 samples per group, adding a jitter point plot is recommended to better estimate data variability.
The paper discusses how the upregulated expression of certain biomarkers correlates with decreased overall survival, contrary to their downregulation in AML versus healthy individuals. A discussion on this paradox would be beneficial.
Additionally, the detailed description of the six biomarkers' functions in previous AML studies is pretty good, and including a sub-figure summarizing this information would further aid reader comprehension.
In this manuscript, the authors performed multiple bioinformatics analysis based on published or public databases with a focus on chemokines. Several proteins were described to be downregulated in acute myeloid leukemia (AML) and were suggested to be involved in patient survival.
The authors claimed that several chemokines or its receptors could serve as biomarkers for AML. But, unless a substantial amount of experimental and clinical evidence being provided, this conclusion cannot be made. Based on current data, “correlation” may be a better term to reflect the involvement of those genes in AML. The title and the main text need to be tuned down accordingly as well.
Protein level of the six putative genes must be analyzed by either western blot or ELISA to further consolidate the downregulation in AML.
Established biomarkers, such as FLT3, NPM1 et al, should be included in key bioinformatic analysis as positive control.
.
.
.
In general, the English writing of this manuscript needs to be improved. I got confused during reading it mostly because there are many long and combined sentences linked with "and" such as line 22-24 or line 52-54. The authors need to re-write the manuscript using more simple sentences to make it clear enough to understand.
The authors conducted a series of analysis using different models. For each panel of all the figures, the authors just told us the analysis they did and the results they got in the main text. I think the readers need some description of the reasons to do certain analysis for each part like line 165-167 to understand all the data.
The authors did a comprehensive analysis to identify the chemokine biomarkers in AML. They also performed some prognostic analysis on the DE genes. However, adding some mechanism discussion on the results will strengthen all the discoveries.
1. In terms of DE genes, how do you think if they are drivers of AML or just biomarkers?
2. For the DE-CCRGs, some improve the overall survival, while some decrease. Are there any hypothesis on these effects?
3. Are there any mutations happening on the CCRGs? Does the differential expression come from mutations or reduced transcription?
For the last part of the manuscript, the authors did a drug interaction network to the DE-CCRGs. Have some of them been used to treat AML? Are there any insights to predict the response of these drugs for AML therapeutics?
All text and materials provided via this peer-review history page are made available under a Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.