HPV Integration Site Mapping: A Rapid Method of Viral Integration Site (VIS) Analysis and Visualization Using Automated Workflows in CLC Microbial Genomics
Round 1
Reviewer 1 Report
It was a pleasure to read this paper by Dr Jane Shen-Gunther and colleagues on HPV integration site mapping. I could not agree more with their statements in the discussion that command-line based, open-source software for increasingly routine bioinformatic tasks must be superseded with simple GUI-based workflows that are accessible for the clinical laboratory.
The paper is well written and makes a significant contribution to this expanding area of HPV research.
I have a few questions and comments:
What was the original sample type that nucleic acids were extract from for the Gabon patient samples? Was it biopsy tissue from colposcopy, liquid based cytology, glass cytology (??) or even self-taken swabs?
Can you please comment, for the reader, on what type of sample would be required for this type of NGS analysis ongoing? This will affect what stage of the screening algorithm the test could be used. For instance, in my country, we will shortly transition to an entirely self-taken swab programme with colposcopy for HPV-16/18 positives and symptomatic women.
I think you need to acknowledge in the discussion that you didnt run any HPV positive but NLM samples in your study. Integration could feasibly precede histological changes and it would be a nice control sample any way. Or has this been done to death already and I just haven't read that paper yet?
I have some comments that I think will enhance the readability of this paper:
1. The two sentences in lines 37 to 41 about the historical role of Professor Harald Zur Hausen, whilst laudable, are not necessary to a paper of this nature. Delete or reword into one short sentence with only 1 reference.
2. On page 2 line 56 you use the word “contradistinction”, which doesn't appear in my thesaurus. I suggest you simply say: “However, the viral genome unwinds bidirectionally...”
3. Is it necessary to reference the statement on line 67-68 about the era of NGS? Isn't it considered common knowledge in this field?
4. Figure 1 should be deleted. It is really hard to understand until the reader has finished reading the paper, the supplementary materials and viewed the MP4 file. It raised a heap of questions for me such as: why was E2 to L2 missing? Why c/s 11? Etc. And since I printed it out in black and white, it was super hard to see.
5. Section 2.3, were you operating in a unix environment?
6. Figure 3 A needs to be converted to a scatter graph or some other sort of figure as the colours dont come out on black and white printing and even when viewed in colour, there is not sufficient difference between the types to be useful. My red-green colour-blind colleague was massively confused and surmised that everything on the graph was significant. In short: it is blinding.
7. Figure 3B, why do you need a map of Gabon on the figure? Surely most of your readers have adequate geographical knowledge?
8. Figure 4 and text in 3.2, please give a short explanation why E2 to L2 is persistently deleted in integrated viral genome sequences.
9. Figure 6. Is there any trend in the location of integration in your samples?
10. 3.3 That is a lot of processing power required for 21 samples and in the context of this being used as a screening and triage tool, most pathology labs would need to upgrade their computing equipment. Please comment on that in the discussion.
Author Response
Dear Reviewer,
Please see attached file.
Thank you so much,
Jane Shen-Gunther
Author Response File: 
Author Response.pdf
Reviewer 2 Report
Sir,
I have reviewed the manuscript "HPV Integration Site Mapping: A Rapid Method of Viral Integration Site (VIS) Analysis and Visualization Using Automated Workflows in CLC Microbial Genomics" submitted by Jane Shen-Gunther and co-workers to IJMS.
First, I was asking myself whether IJMS is the optimal target journal for this scientific communication. It is profoundly based on and also focused on biostatistics. There is actually very little - if any - biology or cancer biology research presented. However, I have concluded for myself that the authors can present their work as proof of concept, which is offered and available to biologists and mainly clinicians for their consideration.
The Introduction is concise and quite clearly written. The authors are right that next-generation sequencing has transformed viral discovery and meta-genomic research. However, I believe that the authors should here formulate a more clinically relevant working hypothesis. Integration of viruses is a biologically interesting phenomenon, indeed. But it does not necessarily cause a clinically apparent disease. If so, I believe that more emphasis should be placed on the potential clinical application in the future.
The material was a publicly available dataset of 21 liquid-based specimens with atypical cytology (all HPV positive). However, this does not include any normal control samples (normal cytology, HPV positive or negative). As HPV infection is very common, and the majority of them can be easily controlled and consequently eliminated by the immune system. therefore, one must very carefully consider what would be real-life utility of this approach and the significance of a positive result. The authors should consider and discuss whether this might be valuable in e.g. borderline cases of uncertain significance (ASCUS).
Further, all the work was done on liquid-based specimens, which is obviously a very convenient material. However, I am just wondering whether the proposed approach would or would not fit more to the FFPE type of material. Viral integration is intimately associated with cancer. The information regarding the nearby genes (as presented in Table S3) would (at least in my eyes) fit more to this FFPE-based application for purposes of patient risk stratification. I believe that the authors could more explicitly discuss this potential.
Further, I miss here any commentary on tumour heterogeneity. I believe that the value of the proposed analytical procedure would be greatly enhanced in association with single-cell sequencing methods. At least, this should be briefly discussed.
To conclude, I believe that the presented work holds certain potential and I believe it can be perfected soon.
Author Response
Dear Reviewer,
Please see attached file.
Thank you so much,
Jane Shen-Gunther
Author Response File: Author Response.pdf
Round 2
Reviewer 2 Report
Sir,
I have studied the resubmitted Manuscript and the Rebuttal letter provided by the authors with the utmost interest.
I am grateful for all the necessary changes made by the authors in the new version. I believe that all those clarifications were at least somewhat helpful and can potentially increase the attractiveness of the described approach to the readers.
I believe that all the answers are now more than satisfactory, scientifically sound and honest.
To conclude, I believe that this manuscript could be considered for publication in IJMS.
