Single-Cell Transcriptome Atlas of Leaves at Different Developmental Stages in Populus alba × Populus glandulosa Clone 84K

Round 1

Reviewer 1 Report

Comments and Suggestions for Authors

The manuscript authored by Jing et al presents a single cell transcriptome analysis of the tender and functional leaves of poplar 84k. The authors have identified distinct clusters for the cell populations and further constructed pseudotime ordering and trajectories. Differentially expressed genes (DEGs) related to specific functional pathways were also identified. I found it is difficult to further evaluate the manuscript as clarifications are needed for some analyses results and the quality of figures are too poor. I suggest the authors carefully revise their text and figures and submit an improved manuscript. Below are some of my comments for consideration. Note that there could be more issues than what has been listed below, I can only provide further evaluation after a manuscript of better quality is received.   

Major:

1.      Quality of all figures need to be further improved. Fonts/text are too small and blurry. Many figures are completely not readable. The figure legends do not clearly explain what the colors represent in figures (e.g. figure 2e, 3e, 4e, 4j, 5e, 6d, 6e, 6f, 7d, 7e, 7f)

2.      The tool used for constructing trajectory (monocle2) relies on a strong assumption that all cells originated from a single source (as in fig. 6) and the cell trajectories form only a single tree with a single root node, which might not be always the case. Unless the authors can provide the evidence (either biological or experimental) showing all cell populations were differentiated from the same progenitors, the use of monocle2 might be inappropriate. The authors also need to explain how they picked the root nodes for monocle2 trajectory trees. Monocle3 can model multiple trajectory trees while not relying on this assumption.

3.      While the analyses could be interesting, different topics seem to be not well connected. The authors should expand the discussion on how functional pathways are different/similar among the different developmental stages and different cell types. This could be an important message to the readers. Currently the pathways and cell types are only discussed separately. An equally important question is whether these observations/conclusions can relate to previous literature?

Minor:

1.      Was UMAP performed on principal components (PCs) ? If so, how many PCs were used? These parameters need to be clarified in the manuscript.

2.      For pathway enrichment analyses, whether p-values were corrected for multiple test comparisons? For the p-values shown in fig 2 and 3, it is hard to see which pathways are significant (p-values < 0.05) because the smallest unit in the legend is 0.25, which is much higher than the 0.05 threshold.

3.      There is no need to elaborate all gene names for every functional pathway as it digresses from the main message of the paper. It's better to list all gene names in supplementary table and only name a few most important ones in the main text.

4.      The authors used subgroups in fig 6 and fig 7 but did not clearly explain how these subgroups were derived. Due to the poor quality of the figure, I don’t know where to find the subgroup information in these figures.

Comments on the Quality of English Language

English writing is good.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 2 Report

Comments and Suggestions for Authors

Jing et al, present an interesting insight into single cell transcriptome atlas of Poplar leaves. They used scRNA-seq technique to how leaves development occurs in Poplar and defining the trajectory of cell differentiation and leaf development. 

 

The study is by and large done well and would be a good resource for the plant science community. But I have few comments for the authors and if they can provide some clarifications:

 

1. Line 40-41- authors talk about different tissue/cell types of Arabidopsis for which the scRNA-seq data is available but they missed the most important cell type that is pollen. There are studies that are done in plant reproductive cell types that should be cited here too: 10.1016/j.celrep.2022.111699 and https://doi.org/10.1007/s00497-018-00355-4

2. Line 41-44: authors have alluded to different scRNA-seq in different plant species without citation. They should provide citations for all those species that they mention. Same for any other species that they mention through out the text like peanut etc. 

3. Line 66-67: authors used tender leaf and functional leaves for their analysis but provide no rationale for doing so or any method to classify them. It would be useful addition for the readers and anyone wanting to repeat. 

4. Can the authors clarify if the leaves they took for the analysis came from same plants or different plant ? and do authors anticipate any biological variations between the plants and the leave age, size that they used in this study

5. Authors have used abbreviation for different cell types like EC, GC, PMCs etc. I would suggest using their full name in the subtitles at least so that readers can follow for eg in 3.2, 3.3 and so on. 

6. Section 3.7 and 3.8 talks about developmental trajectories of tender and functional leaf cell type. But for me, it was very hard to understand the text. Could authors simplify the text and try to bring out main points like what was done, why it was done and what were their main results. Based on their results what were their conclusions and if they see any differences in the developmental trajectories between two cell types

General Comments:

 

Though the manuscript is largely well written but I think the amount of redundant text makes it very difficult to follow what the actual results and conclusions are. The readers would benefit from simplified text overall.

 

Also, figure 2-7 are very poorly done in terms of resolutions. I could not understand most of the panels in those figures. I would suggest to improve the resolution of the figures.  

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

For Figures 2, 3, 4, 5 in a,d,f,i what do you mean with “color” in the legend? p-Value or QValue? The quality of the images could be improved, in particular Figures 2b and 2e. Probably in the final paper will be in a better resolution; the font of the text in some cases are little and the dpi low.

In Figure 2d, 3d, 4d and 4i you bring result of N-L_up and down regulated genes; why you don’t talk about these results in the result section?

enrichment, gene size of the cluster (numbers of genes in the cluster) and p-Value/FDR of the cluster are 3 different parameters in a GO/KEGG analysis (see https://doi.org/10.1021/acs.est.2c01206).  in Figures 2,3,4 what do you reported?

40: You can add also this reference DOI: 10.1016/j.devcel.2022.01.008

41-44 Appear that some references are missed. Add or re-write this paragraph to correlate better the cited references.

51 “are” instead of “is”. Add reference.

53 Add reference

112-121 Could be described better to help understanding the result. See https://doi.org/10.1073/pnas.2018788117

137-138 In figures S2-S5 recall in the caption the parameter used to select the term (like FDR<=0.05 or QValue). You can also specify better in the MM part the use of KEGG

133-147 explain better in with way you find the marker genes. You used the UMAP algorithm and the FindAllMarkers function but you didn’t describe this passage; GO analysis and KEGG can’t bring to this result like appear from the text. Here you need to reference to the Table S3 and S4

142 Why cluster 0? You can’t start from 1

148 In Figures 1a and 1c not so visible numbers. Also, the scale bar in Figures 1c, 1d and 1e could be improved (bar and numbers). Add scale bar in Figures 1f and 1g. Figure 1i appears horizontally stretched.

160-184 you write “Clusters 2 and 3 were mainly enriched for genes related to fatty acid biosynthesis and metabolism, stress resistance, and respiratory metabolic.” From Figure 2A no genes are detected in fatty acid biosynthesis for cluster N2 and N3. Fatty acid metabolism is a more generic KEGG pathway that include fatty acid biosynthesis, elongation, degradation and other. Stress resistance and respiratory metabolic aren’t KEGG classes. Review your results because is a recurrent error (also in paragraph 3.3) and if you group some KEGG categories in a more generic category give the list in brackets. You also grouped clusters N6, N7 and N8 for same KEGG pathways but cluster 8 isn’t so similar to the others two.

197 In Figure 1d you bring result of N-L_up and down regulated genes; why you don’t talk about these results in the result section?

198-207

240 from Figure 2d emerge that there aren’t N-L_up-regulated genes; why you don’t talk about this result in the result section?

Paragraph 3.7-3.8: It is very difficult to follow. The English is good, but it is difficult find the correspondence between the text and the Figures due to the low quality of the Figures in particular when you try to describe groups, subgroups and the different leaf tissue. Improve and try to explain better the result.

I’m sorry but I will not continue to read the discussion and conclusion parts for gaps that in the results due to difficult understanding.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Round 2

Reviewer 2 Report

Comments and Suggestions for Authors

The authors did a good job in improving the manuscript based on the feedback. I hope they can still work on some of the figures as the quality of the figures is still bad.

Comments on the Quality of English Language

Some improvement is required.

Author Response

Please see the attachment.

Author Response File: Author Response.pdf

Reviewer 3 Report

Comments and Suggestions for Authors

Dear Authors

Regarding the answer to my comments concerning the difference between the KEGG pathway founded and the more generic classes attribution that you give to a group of this KEGG classes (for example that most of the genes in the three pathways of Plant-pathogen interaction, Protein processing in endoplasmic reticulum and MAPK signaling pathway-plant are related to stress resistance) now is clear. You can improve this explanation to all the readers, if you add in the S10, S11, S13, S14, S16, S17, S19, S20, S22, S23 a new column in with you make a correspondence between the founded KEGG classes and your generic classes attribution. I hope to be clear.

Figure 7: Regarding the branch points, if the maturation of the cells start from the left high arm, why the first signed branch point is signed with 2 and not with 1? Or is only a graphic issue because the two branching points are very near?

Author Response

Please see the attachment.

Author Response File: Author Response.pdf