Detection of Genomic Uracil Patterns
Round 1
Reviewer 1 Report
The humongous manuscript about the effects of anticancer drugs on uracil in genomic DNA is somewhat unusual; it combines the format of a review and an experimental paper (or papers). The review parts are written with varying depth and expertise (we noticed several errors in terminology, see below). The results section describes biotin ΔUNG sensor; mutation signatures resulting from drug treatment of HCT116 mismatch repair-deficient cell line (is it ultimately relevant to focus of the study?); gene-specific U-score as a new metrics (the most exciting part of the study). Most of the experimental work is descriptive, and the discussion fails, contrary to the author's claim in the title, to provide mechanistic explanations of the author's observations. For example, the levels of APOBECs in treated cultures could have been estimated. This would clarify the controversial issue of correlation of deaminase expression with genomic uracil levels PMID: 30348839. Instead of discussing the obtained results, the authors provide a lengthy discussion of NGS-based applications' methodology. The reviewer's vision is that the manuscript should be split into several separate papers, which different sets of reviewers should review.
Comments:
The title is quite eclectic and long; the message is unclear.
The Abstract needs serious work. It should be succinct and clear. Instead, the authors tend to overexplain many well-known concepts, which makes the piece difficult to read. On the other hand, some authors' claims could not be taken for granted (e.g., the stochastic nature of uracil deamination in DNA and uracil incorporation from the nucleotide pools). A few sentences are complicated but generic, e.g. “The physiological and patho-physiological status (definition?) of the cell regulates the activity of the different deaminases, which possess altered sequence environment preferences (what is sequence environment?) for the cytosines to be deaminated”, or simply wordy without much meaningful information “Recently, the significance of genomic uracil in cellular processes has been established by experimental evidence in numerous publications, and several methods became available for determination of genome-wide uracil patterns”.
Only three last sentences relate to the actual summary of the manuscript. The description should be expanded. It would be good to explain what kind of "mechanistic insights" the authors are looking for.
Line 38. Is it the most frequent? How do deamination rates compare to depurination, oxidation, etc.? Also, “on the other hand” looks unnecessary in the following sentence.
- The selection of references [4-7] for deaminases looks random, except for the excellent Bhagwat’s review. Seminal papers of T. Honjo, M. Neuberger, R. Harris are neglected.
- It is not clear from the text if the polymerases need pool perturbation to start dU incorporation or the perturbated pools provide abundant dUTP substrate?
43-44. DNA context?
In Fig. 1, it would be good to show initial correct CG or AT pairs at different locations. Otherwise, it is confusing to see UG and UA at the same location in row 2. The authors should try to find a better term than misleading “influenced protein binding properties”. In the legend, there is a wrong symbol for one polymerase (correct Pol ζ is substituted by non-existent Pol ξ).
- Remove “in several publications”.
- What is “another particular”?
- Mistake, translesion polymerases, not translation polymerases.
- Pol β cannot be regarded as high fidelity as Pols δ and ε.
130-132. The concluding remark comes out of surprise; the previous text did not address the efficiency of therapies.
- What is “cellular context”?
- Decipher “As such, 5FdUR reports better on TS-targeting effects…”.
- NGS per se is not a tool to study DNA repair, but numerous technique variations are.
- Are the authors sure that “formidable” is right here?
610-611. The concluding remark ruins the elaborate description of methods, plainly stating that "different things" should be considered.
676-677. The authors should discuss in more depth the validity of estimation of mutation frequencies without identification of mutations.
Author Response
Please see the attachment.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript gives an overview of the biological significance of uracil in genomic DNA and describes the techniques and procedures for analyzing the presence of uracil in DNA. Importantly the authors then describe a new technique which they have developed to localize genomic uracil and use this new technique to look at the effects of certain drugs known to dysregulate uracil biochemistry on the organization of uracil in the genome. The strategy appears to be robust and the results that are presented are convincing. As part of their discussion the authors also describe important considerations for analyzing the data generate from arrays and uracil-seq procedures. The studies are an important contribution to the field and the results appear suggest that the location of uracil introduction can be dependent upon the specific pathway that a drug inhibits uracil biochemistry.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Reviewer 3 Report
In this manuscript, Békése et al. have provided a review of methodologies to measure U-DNA within the genome, with a specific emphasis on next generation sequencing (NGS) based-approaches. They then describe a novel U-DNA sensor and a new metric that they refer to as the U-score.
I find both aspects of the paper quite interesting, however, as the authors state in lines 224 to 223, the manuscript has currently been prepared as a combination of two distinct components 1) a review of current methods to measure U-DNA, and 2) a methods paper for a new U-DNA sensor. The rationale for this decision is not clear to me, especially as both components of the paper comprise well written, complete and distinct articles on their own. Furthermore, in my opinion, there is currently too much background for a methods/research paper and too much data for a review article. It is in my opinion that this work should be divided into two separately articles.
Remarkably, I however have few other negative comments. The manuscript provides an excellent review accompanied by informative figures (although I did note in Line 116 that “translation synthesis” should be “translesion synthesis”). The new methodology is also well described and no doubt useful to the field.
Author Response
Please see the attachment.
Author Response File: Author Response.pdf
Round 2
Reviewer 3 Report
I believe the manuscript is now much improved as a stand alone review article. I recommend the manuscript be accepted in present form.
I wish the authors best of luck with publication of their experimental data.
