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.
We appreciate your attention to detail regarding the prior reviews and this manuscript is now ready for publication.
[# PeerJ Staff Note - this decision was reviewed and approved by Brenda Oppert, a PeerJ Section Editor covering this Section #]
The revised manuscript now exemplifies good basic reporting following comprehensive integration of peer review feedback. Noteworthy improvements include the adoption of clear, unambiguous, and professional English language throughout the manuscript, ensuring enhanced readability and comprehension. Additionally, the figures have been refined to uphold relevance, quality, and clarity, with meticulous labeling and comprehensive descriptions that effectively support the findings presented. These enhancements collectively underscore the manuscript's commitment to meeting standards of basic reporting in PeerJ.
The revised manuscript has thoroughly addressed all initial concerns especially with batch effects. The research question is now clearly defined, relevant, and meaningful, with a clear explanation of how the study fills the identified knowledge gap. The investigation has been conducted rigorously, adhering to high technical and ethical standards. Additionally, the methods section has been enhanced with sufficient detail and information, ensuring the study can be accurately replicated.
The findings in the manuscript are promising and can be explored with larger datasets. This study delved into the involvement of macrophages in the inflammatory response following acute myocardial infarction (AMI) by leveraging public datasets from the Gene Expression Omnibus (GEO) database alongside a mouse model. Through analysis of two bulk RNA sequencing and one single-cell sequencing datasets, the research identified differentially expressed genes (DEGs) linked to AMI and further substantiated these findings using another bulk sequencing dataset and a mouse model. The study encompasses a total of only nine training samples each in the AMI and sham groups. Additionally, the validation set comprises only six samples each in the AMI and sham groups, drawn from both the sequencing dataset and the mouse model.
The revised manuscript demonstrates significant improvements in basic reporting. This manuscript can be published based on its validation from independent bulk sequencing dataset and the mice model, as they reinforce the biomarkers identified from training dataset.
In addition to the two reviewers, I reviewed the paper and shared the extensive concerns raised with reviewer 1. All issues raised need to be fully addressed for the article to be suitable for publication, particularly concerns around data accessibility, experimental (including computational, like, for example, are version numbers, settings, etc.) details necessary for reproducibility, and the statistical and analytical concerns.
This study delved into the involvement of macrophages in the inflammatory response following acute myocardial infarction (AMI) by leveraging public datasets from the Gene Expression Omnibus (GEO) database alongside a mouse model.Through analysis of two bulk RNA sequencing and one single-cell sequencing datasets, the research identified differentially expressed genes (DEGs) linked to AMI and further substantiated these findings using another bulk sequencing dataset and a mouse model. The study encompasses a total of nine training samples each in the AMI and sham groups. Additionally, the validation set comprises six samples each in the AMI and sham groups, drawn from both the sequencing dataset and the mouse model. Publication’s introductory sections provide essential context, ensuring readers grasp the significance of the topic from the outset. Following are the issues found with basic reporting:
1) Figure 2-A and B: Please enhance the font size of the labels as they are currently challenging to read.
2) Figure 3 Legend title is misleading since there is no functional enrichment analysis in Figure 3.
3) Figure 3A: The log2 fold change threshold specified in the manuscript text for this analysis is 0.5. However, it appears unclear whether the dotted line in this plot aligns with this threshold.
4) Figure 3B: This visualization doesn’t augment the context provided in the existing manuscript text concerning this finding. Instead a plot demonstrating Co-DEGs exhibit similar fold changes in both single-cell and bulk RNA-seq data, both in terms of quantity and direction would be informative.
5) Figure 3C: The figure legend and the figure doesn’t specify expression levels of which samples are shown.
6) Figure 5A: The Y-axis label doesn’t clearly indicate the represented value.
7) Figure 5B: Color legend of the plot doesn’t specify the represented value. It is also quite unclear from this plot how the key screening genes were selected based on the coloring scheme. For example: Why not select Thbs1 as well?
8) Figure 5C: Color legend of the plot doesn’t specify the represented value.
9) Figure 7: The figure legends suggest that these plots validate the discovery of the four key genes. However, the term 'validation' regarding the four key genes in GSE163465 and GSE236374 datasets is misleading, as these datasets are part of the training data.
10) Figure 7-E, F, G, H and Figure 8-A, B, C, D: It is unclear from the figures what value is represented by the y-axis.
11) Figure 8-E, F, G,H: The annotated text within the plot is quite difficult to read.
12) Figures 9, 10, 11: Just a suggestion to maintain consistency with other figures: consider displaying the actual p-value in the plots.
13) Line 80: Correct identifier for GEO dataset is GSE114695 instead of GSE1146945.
14) Line 112: “The screened key genes were subjected to Pearson correlation analysis with genes in inflammatory response-related pathways to analysis the degree of correlation between key genes them and inflammation-related pathway genes” can be reframed to “The screened key genes underwent Pearson correlation analysis with genes in inflammatory response pathways to assess the extent of correlation between these key genes and genes associated with inflammation-related pathways” for better clarity.
15) Line 140: “threatment” should be “treatment”.
16) In the RT-qPCR.xlsx file extracted from the raw data, sample identifiers are absent from the columns.
17) Could you please provide the gene expression data files for RNASeq used to generate the figures and analyses presented in this article? Sharing this data would enhance transparency and facilitate the reproducibility of your work.
Overall this study presents a nicely crafted research question and by explicitly outlining how the research bridges an identified knowledge gap, the publication underscores its relevance and potential impact, making a good case for its utility in advancing the field's understanding. But here are some of the issues with methodologies in the study:
1) While the research question is well-defined, it is worth noting that the size of both the training and validation datasets is small. This limitation underscores the need for cautious interpretation of the findings, as the sample size may affect the generalizability and robustness of the results. Despite this constraint, the study offers valuable insights within the scope of its available data, highlighting areas for potential future research with larger datasets to further validate and extend its findings.
2) This study has utilized public RNASeq datasets from four different sources with varying assay workflows and protocols. This diversity introduces a significant concern regarding data comparability and consistency across studies. The use of different laboratory techniques and sequencing platforms can lead to systematic biases and confounding factors in the analysis. As a result, one must exercise caution when integrating or comparing data from multiple sources, as differences in experimental protocols may obscure true biological signals or generate misleading conclusions. Addressing these concerns typically involves robust normalization and quality control measures to minimize the impact of technical variability and ensure the reliability and reproducibility of downstream analyses. I noticed following concerns in the methodology deployed by the authors:
a) According to the data retrieved from https://www.ncbi.nlm.nih.gov/gds/?term=GSE236374[Accession], dataset GSE236374 consists of two replicates for each of the three AMI samples. One replicate corresponds to day 7, while the other corresponds to day 28. However, the manuscript does not specify which replicate was utilized in this study.
b) The authors have omitted detailed descriptions of the bioinformatics pipeline utilized for processing bulk RNASeq datasets. For instance, clarification is needed regarding the alignment procedure, including the choice of genomic reference. Furthermore, it is unclear whether any quality control filtering or contamination checks were conducted.
c) It appears that the bulk RNASeq datasets GSE236374 and GSE183272 were merged for the purpose of conducting differential expression analysis. However, the metadata does not provide clarity regarding whether these studies include samples from the same day, employ the same assay workflow, or have consistent sequencing depth.
d) The authors state the utilization of the limma package for normalizing bulk RNASeq datasets to mitigate batch effects. However, it remains unclear whether this choice effectively addresses all variances introduced by the two distinct sources of datasets mentioned previously. Furthermore, the specific function employed from the package has not been disclosed.
e) It is unclear if any normalization was done with the validation dataset GSE114695.
3) Lines 109 - 113: The authors used GSVA algorithm to identify key genes from Co-DEGs set based on their correlation to inflammation pathway genes. It is still unclear how specifically those 4 genes were selected. For example why not consider THbs1? Was there any existing overlap between Co-DEGs set and inflammation-related pathway genes?
4) Lines 114 - 117: This section is perplexing as datasets GSE163465 and GSE236374 are included in the training samples. They were utilized in the discovery of differentially expressed genes, hence should not be employed for validation purposes.
In this study some of the conclusions are supported by findings of the research, ensuring a clear and logical connection between the study's objectives and its outcomes. But following are the issues with the results section:
1) Lines 204 - 208: The authors have not specified whether they checked for direction of change as well while intersecting genes between differentially expressed genes of bulk and single cell RNASeq.
2) Lines 219 - 224: The process by which the four key genes were chosen remains unclear, leaving ambiguity as to why certain genes were prioritized over others. Updating the figures would enhance clarity and comprehension of this finding.
3) Lines 229 - 234: This section is concerning since training datasets are not suitable for validation purposes.
The overall structure of the paper is well-organized, but there are certain sections where clarity could be improved.
1. The manuscript is well-written in clear, professional English, suitable for a scientific audience. No major language or grammatical issues are noted that would impede understanding.
However, a few minor improvements could be beneficial:
(1) GSE183272obtained in line 25 need a space
(2) line 191: comma
(3) Unclear sentences start from line 122, missing period: "The screened key genes were subjected to Pearson correlation analysis with genes in inflammatory response-related pathways to analysis the degree of correlation between key genes them and inflammation-related pathway genes"
2. The manuscript's structure conforms well to the standards expected by PeerJ, following a clear and logical format typical of scientific papers.
3. The figures included in the manuscript are relevant, high-quality, and contribute effectively to illustrating the study's findings. Each figure is well-labeled with detailed legends that adequately describe the contents and significance of the visual data.
However, some figures are suggested to adjust the format without distortion, such as Figure 1E, 3C.
1. The manuscript is focused on identifying macrophage-associated inflammatory genes in acute myocardial infarction (AMI) using a combination of single-cell and bulk RNA sequencing. Given the detailed analysis and relevance to inflammatory processes in a critical medical condition like AMI, the research aligns well with scientific inquiry in cardiovascular pathology and molecular biology. The manuscript appears to be original primary research suitable for a journal focused on these areas.
2. The research question is clearly defined and centers on discovering key macrophage-associated genes involved in the post-AMI inflammatory response. The study addresses a significant knowledge gap in understanding the molecular mechanisms of AMI-induced inflammation and the role of macrophages, providing valuable insights that could lead to better therapeutic strategies.
3. The investigation seems rigorous, combining advanced bioinformatics analysis with experimental validation. The methods section details the datasets used, analysis techniques, and validation methods, ensuring a comprehensive approach to the research question.
4.The methods are described meticulously, providing enough detail for replication.
1. The manuscript does not explicitly assess the impact and novelty of its findings within the broader field of cardiovascular research, which is a slight drawback. However, it does emphasize the potential therapeutic implications of the identified genes in AMI, suggesting that these findings could contribute significantly to understanding and managing post-AMI inflammation.
2. The manuscript provides a detailed description of the data sources (e.g., specific datasets from the Gene Expression Omnibus), the analytical methods used (including bioinformatics tools and statistical analyses), and controls applied in the study. The data encompass single-cell and bulk RNA sequencing analyses, which are complemented by rigorous statistical methods (e.g., differential gene expression analysis, gene set enrichment analysis).
3. The conclusions are explicitly linked to the original research question concerning the identification and characterization of macrophage-associated genes in AMI. The findings discussed in the conclusion directly relate to the experimental and analytical results presented, such as the identification of specific genes linked to inflammatory pathways and their validation in AMI models.
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.