Alternative Splicing (AS) Dynamics in Dwarf Soybean Derived from Cross of Glycine max and Glycine soja
, Rahul Vasudeo Ramekar 1,2, Taeyoung Um 1,2, Ju-Kyung Yu 3, Kyong-Cheul Park 1,2,* and Ik-Young Choi 1,2,*
Round 1
Reviewer 1 Report
The revisions have not addressed most of the concerns raised in the first review. The revised manuscript still has not shown that sufficient replications had been conducted and properly integrated in the data analyses. There is still no corroboration using established or alternate analytical methods, nor more in-depth look on the effect and magnitude of each AS type on the genes of interest.
Author Response
Response to Reviewer 1 Comments
The revisions have not addressed most of the concerns raised in the first review. The revised manuscript still has not shown that sufficient replications had been conducted and properly integrated in the data analyses. There is still no corroboration using established or alternate analytical methods, nor more in-depth look on the effect and magnitude of each AS type on the genes of interest.
Response 1: Thank you for your comment. We agree we could not add any additional analysis considering the sequencing cost and the resource limitation we had. However, we have mentioned that the manuscript provides comprehensive information on the dynamics in alternative splices occurring in closely related plants with different phenotypes. This data will be helpful in designing and further understanding the roles and molecular mechanisms of AS.
Reviewer 2 Report
The authors incorporated all corrections nicely. No further corrections are required.
Please revise the MS title as "Alternative splicing dynamics in dwarf soybean derived from the cross of Glycine max and Glycine soja"
Author Response
Response to Reviewer 2 Comments
The authors incorporated all corrections nicely. No further corrections are required.
Please revise the MS title to "Alternative splicing dynamics in dwarf soybean derived from the cross of Glycine max and Glycine soja"
Response 1: Thank you for your comment. We agree with the suggestion and we have incorporated the revision of MS title on Page 1 highlighted.
Reviewer 3 Report
1. In the abstract in line 25, "As a result, 832 and 36,772 nonredundant transcripts were generated"? 34,832, not 832, nonredundant transcripts were generated actually.
2.Is this a revised manuscript? Why so many modification traces in the manuscript?
3. Why Only 15,570 and 16,608 genes were identified in dwarf and normal soybean in this manuscript? Up to now, more than 50,000 genes were predicted according to the reference genome.
Author Response
Response to Reviewer 3 Comments
- In the abstract in line 25, "As a result, 832 and 36,772 nonredundant transcripts were generated"? 34,832, not 832, nonredundant transcripts were generated actually.
Response: Thank you for noticing. We have that 34,832 in the previous versions and it somehow got missed while submitting the last version. We have corrected the mistake in the abstract.
2.Is this a revised manuscript? Why so many modification traces in the manuscript?
Response: Thank you for your comment. The revised manuscript has modifications incorporated by all the reviewers. But the main idea didn’t change throughout the MS.
- Why Only 15,570 and 16,608 genes were identified in dwarf and normal soybean in this manuscript? Up to now, more than 50,000 genes were predicted according to the reference genome.
Response: Thank you for so interesting comment. The genes identified were 15,570 and 16,608 in dwarf and normal lines. The reason behind a smaller number of genes detected by RNA-seq can be attributed to the fact that it was not expressed in the leaf tissue. The reference genome used the all-different tissue types to predict the total number of genes, whereas we have used transcriptome which identified only those genes that were expressed at a particular stage. Similarly, there is another possibility of having a very low expression of other genes and hence they were filtered out and consequently not identified in our study.
Round 2
Reviewer 1 Report
The revisions have not addressed most of the concerns raised in the first review. The revised manuscript still has not shown that sufficient replications had been conducted and properly integrated in the data analyses. There is still no corroboration using established or alternate analytical methods, nor more in-depth look on the effect and magnitude of each AS type on the genes of interest
Author Response
Thank you for your comment. We already responded to your comment in the first revision file with that We agree we could not add any additional analysis considering the sequencing cost and the resource limitation we had. However, we have mentioned that the manuscript provides comprehensive information on the dynamics in alternative splices occurring in closely related plants with different phenotypes. This data will be helpful in designing and further understanding the roles and molecular mechanisms of AS.
Reviewer 3 Report
The author responded to the reviewers' opinions and improved the manuscript, suggesting that it be accepted for publication.
Author Response
We have made the English editing from a professional service provider. We are grateful for your comment and look forward to the manuscript's publication.
This manuscript is a resubmission of an earlier submission. The following is a list of the peer review reports and author responses from that submission.
Round 1
Reviewer 1 Report
The manuscript by Roy et al. reported the splicing can change the growth and defense-related genes from dwarf and normal soybean lines derived from the cross of Glycine max var. Peking (G. max) and G. soja var. IT182936 in an F7 RIL population to study the differences between isoforms. The experiment is well designed/ presented, and the results are of value, but in its current form, the manuscript is not acceptable for publication. The primary shortcoming lies in the week presentation of results. The authors are requested to attend to the following matters and submit a revised manuscript.
- Line 2-4: The title should be changed
- Line 15-31: Abstract needs improvement
- Line 35-36: Plz revise the sentence
- Line 45-59: Need improvement
- Line 109-124: Plz revise carefully, some sentences meaning not clear
- Line 139: Change heading (2.3)
- Line 284-354: Discussion needs improvement
- The manuscript abounds with grammatical errors which should be attended to.
Reviewer 2 Report
The manuscript presented an interesting investigation of the role of alternative splicing (AS) in regulating dwarfness phenotype in soybean. The research has novelty since it utilized long-read PacBio RNA sequencing data, which can provide new perspectives compared to AS studies that rely on short-read next generation sequencing data. However, the manuscript in its current form lacks convincing arguments and evidences to back up its claims, making it not yet suitable for publication in a journal like Agronomy. Some critical points that need to be considered:
1. The authors did not address the possibility that a significant portion of AS events that they detected are simply the product of noisy splicing events (eg. see Wan and Larson 2018, https://doi.org/10.1186/s13059-018-1467-4), whose products are not likely to have any impact on the trait that they studied. The existence of noisy splicing means that numerous AS variants will be found as long as sufficient RNA sequence data are examined, and the challenge lies in identifying the true signal among the noise. Ideally, efforts should be made to ensure that the study has sequence data with sufficient depth to discriminate consequential isoforms from basal noise, as well as sufficient biological replicates to identify isoforms that are consistently present in samples from different individuals with the same phenotype. The use of transcripts from a single tissue from a single representative individual for each phenotype group is thus insufficient for this type of study. If such steps cannot be done due to cost constraint, then at least the authors must acknowledge the limitation of their study in addressing this issue.
2. The use of long-read sequence data in this study is advantageous since the full sequence of many isoforms becomes available, which means that the impact of AS on the composition of expressed proteins can be ascertained. While PacBio sequences tend to have higher error rates, some software packages can correct them using short reads data, which the authors possess (as stated in the introduction on page 3 line 99). So instead of speculating whether some isoforms create premature STOP codons or produce non-functional proteins, the authors should show with confidence which isoforms actually bear such alterations. The authors also missed the opportunity to do a comparison of AS identification using long reads and short reads in soybeans, which can also be used as a validation step for AS isoform identification. It will also be a valuable information for other researchers who are still trying to decide which technology to use in their next experiments involving AS.
3. The candidate gene selection to illustrate possible AS roles on dwarfness in section 3.4 is based on big differences in isoform numbers between the two groups. This may not be a proper selection criterion since high isoform number may not matter much on a highly expressed gene, as long as the default spliced transcript remains dominant. On the other hand, a single isoform at high concentration may have a more significant impact on phenotypes. Thus isoform quantity parameter (FPKM or TPM) should also be considered when selecting the candidate isoforms.
4. The title should be revised as it claims that increased AS were obtained in certain genes of the dwarf soybean, but the data shows that the reverse is true in some genes (figure 6). A contradictory conclusion is also presented in line 272-273 ("We observed that there were more gene isoforms in the dwarf line compared with normal controls") while line 275-276 clearly show that there were less isoforms in the dwarf plant ("the 61 genes produced 247 and 274 isoforms in dwarf and normal plants, respectively")
5. The classification of FSMs (full splice match) as one of the AS isoforms is questionable since their splicing pattern fully match the default splicing. Even if they have variations around TSS and TTS, they are the byproducts of transcription process, not the RNA splicing machinery.
6. The numbers presented in some figures and tables do not seem to match or support each other. For example in figure 3, if the number of genes are multiplied and added (eg. 1 isoform genes * 1 + 2 isoform genes * 2 + 3 isoform genes * 3 + etc etc) the result do not match with the number of isoforms in table 2. Some terms may also need clearer definition, such as the difference between genes and isoforms, and whether there are intersection between them.
7. Writing consistency should be rechecked. For example line 96 stated that dwarfness appeared since F2 and inherited in subsequent generations, but line 114 stated that dwarf F2 plants did not produce seeds. Another example is in line 178, which mentions that the library is a multiple-tissue hybrid library, yet in line 323 it was plainly stated that the sequences were originated from a single leaf tissue.
8. Another aspect that could be interesting to explore is the AS status in the parental lines, to see whether AS pattern is correlated with parental alleles, or if dwarf lines simply produce unique splicing pattern regardless of the allele of the genes being spliced.
