A Novel Proteogenomic Integration Strategy Expands the Breadth of Neo-Epitope Sources
, Le Zhang 3,†, Fanyu Bu 2, Xiangyu Guan 1,2, Lei Chen 2, Haibo Zhang 2, Yuntong Zhao 2, Huanyi Chen 2, Weicong Zhang 1,2, Yijian Li 1,2,4, Leo Jingyu Lee 3, Zhanlong Mei 5, Yuan Rao 5, Ying Gu 2,6, Yong Hou 5, Feng Mu 5,* and Xuan Dong 2,4,*
Luke Maggs
Round 1
Reviewer 1 Report
The authors present an efficient method for identifying neoepitopes from mass spectrometry data. The method is a simple combination of existing tools. The results obtained are not particularly novel. It seems that only additive effects of the algorithms were obtained. The description of the paper seems to meet the technical requirements. On the other hand, it is full of technical descriptions and is not written in an easy-to-understand manner, even to the readers interested in the prediction of neoantigens. Since this is an important issue for the development and improvement of immunotherapy, and since this work seems to make certain contributions to the field, I consider the paper will be acceptable if it is revised to address the following concerns.
Major points
#1
The tool pNovo3 is used in the analysis. However, without reading the original paper of pNovo3, it is difficult to understand what pNovo3 does and what kind of peptides are being identified with pNovo3. Explanation of pNovo3 is needed when pNovo3 is first mentioned.
#2
How accurate are the peptides specific to pNovo3 in Figure 1B? Among the 2346 peptides detected specifically for pNovo3, there may be many false positive candidates. I do not understand why the high correlation between predicted retention time and observed retention time in Figure 2A indicates an accuracy of the identified immunopeptidome.
#3
Regarding the Figures 2B and 2D, it is informative to show the distribution of affinities for MaxCount, pFind3, and pNovo3, respectively.
#4
It would be informative to show what kind of non-canonical peptides were identified. Were they intergenic, upstream, intron, downstream, or read-through?
#5
Did the authors include peptides sequences derived from frameshift mutations into the database of neoepitope candidates?
#6
Were wild-type peptides corresponding to the mutant peptides listed in Table 1 detected?
#7
I am not convinced that one could divide the 9-mer into even smaller segments, say 3-mer and 6-mer, and assign them correctly and unambiguously. I don't think it is possible to correctly identify cis-spliced peptides in the described way. For example, if we randomly extracted peptides with a definite origin and performed the same analysis with a reference that masked the original region, similar number of peptides may be identified as cis-spliced peptides as false-positive results. Alternatively, they could be derived from cryptic splicing not registered in the reference transcriptome. In addition, the authors should explicitly show what kind of peptides were determined to be cis-spliced peptides. Pathway analysis or gene ontology analysis may be helpful to describe the features of identified peptides. It is also required to access whether there were any repeating sequences within the peptides that may cause misassignment.
Minor comments
In the figures (especially, Figures 5D,E), the legends are not clear enough (too small) to see the detail.
Author Response
Please see the attachment.
Author Response File: 
Author Response.docx
Reviewer 2 Report
This is a nicely thought through manuscript detailing a novel method to assess HLA presented peptides in HCT116 cells and identify neo-epitopes.
Major:
· More detail on how the specific strategy designed overcomes the insufficiencies of current database searches and differs from other published analyses of non-canonical peptides in tumor immunopeptidomes would help to highlight the significance of the work.
· The relevance of the strategy needs to be proven either by the detection of novel epitopes in other cell lines or proving the immunogenicity of the neo-epitopes identified in this study.
Minor:
· Some errors in the text need to be checked, e.g. Table 2 is indicated (line 419) rather than Table 1.
Author Response
Please see the attachment.
Author Response File: Author Response.docx
Round 2
Reviewer 1 Report
The Authors have addressed all of my concerns with the original manuscript. I recommend accepting without further changes.
Author Response
We appreciate your helpful comments on our manuscript.
Reviewer 2 Report
Thank you for addressing the comments and providing the additional cell line information. The significance of the manuscript is greatly improved. Mentioning the future need to assess immunogenicity in the discussion could be added. Minor grammatical check of the new text is needed , e.g. line 121; database or databases? line 123; inflate or inflates?
Author Response
Point 1: Mentioning the future need to assess immunogenicity in the discussion could be added.
Response 1: Thanks for your valuable suggestion. We have added the future need to assess immunogenicity in revised manuscript (Line 584-586) as below.
“Besides, the immunogenicity of the canonical and non-canonical peptides identified by our integrative strategy needs to be further validated as neo-antigens for their potential application in immunotherapy.”
Point 2: Minor grammatical check of the new text is needed, e.g. line 121; database or databases? line 123; inflate or inflates?
Response 2: We are sorry for the grammatical mistakes. We have checked the manuscript and made two corrections. For the sentence in Line 90 (Line 121 in PDF version), we added “the” before “database”. And we used “inflates” instead of “inflate” in Line 91(Line 123 in PDF version).
