Transcriptomic Changes Induced by Drought Stress in Hardneck Garlic during the Bolting/Bulbing Stage

Round 1

Reviewer 1 Report

The authors analyse the transcriptional changes that occur in bolting garlic bulbs exposed to drought stress. This is a critical developmental stage, and so this study will provide a useful dataset for studying drought in garlic and related species. The report focuses on several known drought-related genes and pathways, such as ABA biosynthesis, HSPs and E3 ubiquitin ligases, and shows that these are all up-regulated in response to drought. The list of drought-responsive genes is also compared to salt-responsive genes identified separately.

The manuscript is concise and well-written and the analysis pipelines seem sufficient for this type of study. It would be useful to know if any specific non-default settings were used during the bioinformatics steps (or to provide the scripts used). The data also appear to support the conclusions by the authors, although the results are unsurprising. It would have been interesting to have seen a more thorough comparison of this dataset to ones at other developmental stages, or to other Allium species, but there do not appear to be many at this current time.

Major Points:

1. The experimental details and origin of the salt dataset should be available, as well as a reference to the publication from which they are derived. It is currently unclear to me whether the salt dataset (PRJNA682570) was run through the same analysis pipeline as the drought dataset by these authors for the reported comparison (although this is not addressed in the Methods), or whether the gene lists were provided independently (in which case there should be a reference to that study)
2. More clarity on the software parameters used in the bioinformatics pipelines would be useful, or the exact scripts. For example, are the reported/cut-off LFCs with or without DESeq2 LFC shrinkage?
3. Related to the above: was BLAST2GO used for functional annotation as well as GO enrichment? If so, more information is required - e.g. which databases were used for annotation, alignment settings, the use of Annex and imported interpro terms, etc.

Minor Points:

1. The Figure 3B bar plot seems largely irrelevant. The number of genes could be superimposed on Figure 3A.
2. Regarding Figures 3, 4, 5 and 6: all these figures (and to a lesser extent Fig. 8) rely on a green-red colour palette, which would be difficult to interpret for a colour-blind reader.
3. On two occasions (line 199 and 265) it is mentioned that the ABA content increased in the leaves of the drought-treated garlic. While it seems likely that this is true, given the literature, I do not see any data (ABA measurements) to support this specific wording?
4. I cannot find the salt study experimental data (PRJNA682570) or the data from this study (PRJNA683607). They may be under embargo, but if not verify that the codes are correct.
5. I am not certain I understand Figure 7. Is the heat map showing the log2 of the raw (original) read counts in each sample? If so, why not the normalised read counts? Or, if the goal is to highlight the drought-induced up-regulation of these genes, the counts in the D relative to the CK samples?

Author Response

 

Author Response File: Author Response.docx

Reviewer 2 Report

Review of Zhou et al.

Transcriptomic Changes Induced by Drought Stress in Hard-neck Garlic during the Bolting/Bulbing Stage.

The study employed a simplified into the greenhouse experimental design to evaluate the Drought Stress when bolting/bulbing garlic hardneck plants were exposed to water-limiting stress.

The authors expect to extrapolate the results observed in order to explain the adaptations of the plants.

The article is generally well written, however there are a few grammatical errors and lapses in usage that could be improved.

There are also a few issues in terms of interpretation of the data, particularly in the light pervasive expression of genomes that, when addressed, would make the paper stronger.

The results section under the title "Functional categorization of the DEGs" is too long and descriptive.

It presents results about several genes (that are not necessarily functional correlated) in a single (brutally) long text. I would recommend subdividing the text by topics (biological process, metabolic process, and oxidation-reduction process etc., or any other topics that best fits the manuscript).

There are several phrases lacking clarity. Was the text revised by editing experts? Some of them contain important results and comments that should be better explained. Authors do not show statistical analysis for differences observed between control and treatment.

Most of the analyzed compounds display a substantial difference between treatment. Were the differences statically significant? I would recommend including it in the analysis for accuracy. The processing of plants RNAs, is highly complex (see Croucher and Thomsom 2010). RNAs can be processed at the whole strand even gene level using multiple nested promoters.

This can significantly confound the interpretation of RNA-seq. Presenting information on RNA processing in plants would help the non-specialist reader. In particular direct processing of transcripts in plants is important as RNA is directly converted to protein, with control at the ribosome level.

Moreover, as the authors did not employ strand-specific RNA-seq they cannot distinguish between RNAi and coding sequences. An analysis of genes expression and mapping completeness of a given gene might help with this issue. Pervasive transcription is another potential issue, and its potential effect on the conclusions should be addressed. Conclusion is shy. Listing upregulates and downregulated genes is not a conclusion.

I would rewrite it to include major results and briefly discuss the novel aspects of response brought by the study. Also, genes and pathways cited in conclusion are not the same found in the abstract. Why there is such difference?

I believe these two sections should discuss the same major achievements.

Data Availability A major flaw is that the base datasets have not been made available. As the instructions for authors state.

As the interpretation and reducibility of the study is totally reliant on the RNA-seq transcriptomic data it is not possible to reproduce the study without these datasets.

The RNA-seq data should be deposited in NCBI's SRA database or NCBI's Gene Expression Omnibus (GEO).

Author Response

 

Author Response File: Author Response.docx

Round 2

Reviewer 2 Report

Dear Authors,

Thanks for attending all the suggestions. The manuscript has been significantly improved. In view of the above, I believe that the article presents robust and consolidated content, bringing to light new information on changes Induced by drought stress in Allium sativum. I consider that the work has enough quality to be considered for publication in Agronomy (MDPI).