Propagation Fidelity and Kinship of Tomato Varieties ‘UC 82’ and ‘M82’ Revealed by Analysis of Sequence Variation

Round 1

Reviewer 1 Report

The authors compare the cultivars UC82 and M82 based on small sequence variants (SNP array and re-sequencing) to reveal their kinship. Additionally, the authors study the fidelity of tomato propagation.

Major comments:
1) The authors might want to define ‚accession‘ and ‚cultivar‘ in the abstract. Both terms are often used as synonyms in other publications and would probably confuse some readers. In the discussion, the authors also use the term ‘varieties’ which should be defined as well.
2) The abstract is lengthy and hard to understand due to many numbers. The authors might want to rewrite it in a more concise way without any numbers.
3) The authors might want to elaborate on the „SolCAP SNP array“. As this is central for the presented study, details should be presented in this manuscript.
4) The authors might want to check if GATK was really run on mappings with different Phred score scales per sample (Phred33 / Phred64). At least in previous GATK versions, there used to be an error when processing mappings with Phred64.
5) Was no lower coverage cutoff used in the identification of SNPs?
6) The authors might want to include a figure about the different types of SNVs that are distinguished (e.g. polymorphic).
7) The authors could use joint genotyping in GATK to process the four samples in parallel. This would boost the sensitivity of this analysis. Was joint genotyping considered and is there a reason why it was not applied?
8) The largest number of homozygous SNPs was observed on Chromosome 1. Is there an explanation for this observation or is it possible that this is an artifact of the mapping? Chromosome 1 or Chromosome 11 might be the first sequences in the file. If many reads would have been misplaced on these sequences, high numbers of erroneous variants could be expected.
9) Chromosomal positions “bottom” and “top” could be replaced by “north” and “south”.
10) The start of the discussion section (multiple paragraphs) is redundant with the results/method section. The authors might want to write it in a more concise way.
11) The differences between the accessions should be visualized e.g. using a phylogenetic tree. It might be helpful to include an additional data set as some kind of outgroup. Especially to demonstrate high fidelity propagation, a comparison to a very different cultivar is necessary.

Minor comments:
Title: The authors might want to check if two spaces could be removed from the title.
Line 21: The authors might want to rephrase this sentence (“between variation between”).
Line 54: contain > contains ?
Line 117: Which function? Fastq-dump?
Line 215: sequence-based SNPs > SNPs
Line 219 / 220: The authors might want to rephrase the sentence.
Parameters of the PCR duplicate removal could be added.
Line x (line numbers missing): genome > genome sequence
Line x: “artificially condensed” needs to be rephrased. Not the DNA is condensed, but the reference sequence.
Line x: genome reference > reference genome sequence

Author Response

Dear Ms. Rita Ren:

We appreciate the reviewers comments and believe that their questions and suggestions have strengthened the MS. We have addressed comments from the three reviewers (two sent to us via e-mail, and a third which we retrieved from the MDPI portal). To address comments about organization, we have altered the title to provide a less technical and more generally interesting one. We rewrote the abstract to be less technical and reorganized the introduction as well. In order to address the more substantive comments concerning content, we added to a supplemental figure and have now made that figure 1 in the MS. We also added a tree based on clustering to the supplemental files to address a specific suggestion by one of the reviewers. Details and clarification to materials and methods has been in response to specific queries from reviewers. A detail response to the reviewers comments in detail follows.

Thank you for the opportunity to respond.

Yours, David Francis

Reviewer 1

Open Review

(x) I would not like to sign my review report
( ) I would like to sign my review report

English language and style

( ) Extensive editing of English language and style required
( ) Moderate English changes required
(x) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style

Comments and Suggestions for Authors

The authors compare the cultivars UC82 and M82 based on small sequence variants (SNP array and re-sequencing) to reveal their kinship. Additionally, the authors study the fidelity of tomato propagation.

Major comments:
1) The authors might want to define ‚accession‘ and ‚cultivar‘ in the abstract. Both terms are often used as synonyms in other publications and would probably confuse some readers. In the discussion, the authors also use the term ‘varieties’ which should be defined as well.

We have re-written the abstract and provide a definition in the first sentence “Plant varieties are named and released based on distinct, unique and stable characteristics but may be maintained separately by genebanks or stock centers under separate accession identification numbers”. As part of this re-write, we have removed use of Cultivar because M82 and UC 82 are no longer cultivated varieties. This is a simplifying step. We are clear to refer to “accessions” as independently curated samples of a variety.

2) The abstract is lengthy and hard to understand due to many numbers. The authors might want to rewrite it in a more concise way without any numbers.
We have re-written the abstract and removed all but general statements of results rather than numerical descriptions of results.

3) The authors might want to elaborate on the „SolCAP SNP array“. As this is central for the presented study, details should be presented in this manuscript.
The SolCAP array has been described in previous publications. Because our description was not clear to this reviewer, we have re-written the methods to provide a better description in order to address this reviewer’s concern.

4) The authors might want to check if GATK was really run on mappings with different Phred score scales per sample (Phred33 / Phred64). At least in previous GATK versions, there used to be an error when processing mappings with Phred64.

This comment is another indication that our original description may not have been clear, or the reviewer used a different version of GATK. The options “--phred64-quals” and “--phred33-quals” are used in the TrimGalore and BowTie2 portion of the analysis and are appropriate for the sequencing chemistry and Illumina encoding used in the next generation sequencing. Both options correspond to a Phred score of 20, or 99% accuracy.  Use of these options did not affect our mapping using the version 4.0.9.0 of GATK. We are confident that our pipeline used best practices.

5) Was no lower coverage cutoff used in the identification of SNPs?

This comment reveals another instance of where our methods needed to be more clearly stated.  There are two issues with sequence depth of coverage.  The first is caused by insufficient coverage.  We set a minimum of 4x, and did not perform analysis on sequences with less coverage for quality control reasons (there is simply not enough coverage to correct for sequence error). The second issue occurs when too wide a range of coverage is used, and repetitive sequences are mapped on top of the same reference leading to an overestimation of polymorphism, usually heterozygosity. We are also specific about how this upper range was calculated using the method described by Li, 2014 (Li, H. Toward better understanding of artifacts in variant calling from high-coverage samples. Bioinformatics 2014, 30, 2843–2851).

6) The authors might want to include a figure about the different types of SNVs that are distinguished (e.g. polymorphic).

We are not clear on this suggestion.  Figure 1 provides a graphic of Single Nucleotide Variation across the genome (what we think of as google earth scale) while Figure S1 provides a close view for context. We modified the Figure S1 to present a schematic representation of robust SNPs and those that are due to either sequencing error or alignment of multiple genes. You can find it in the new version of the manuscript named as Figure 1.

7) The authors could use joint genotyping in GATK to process the four samples in parallel. This would boost the sensitivity of this analysis. Was joint genotyping considered and is there a reason why it was not applied?

This suggestion describes exactly what we did. Three different approaches were followed to perform whole genome comparison among the four sequence samples.

1. As a first approach we compared the four samples in pairwise combinations, so six different VCF files were obtained.
2. The second approach resulted in a single VCF containing all four samples. The function GenotypeGVCFs in GATK version 4.0.9.0 [28,29] was used to perform joint genotyping for all samples included in each gVCF, resulting in Variant Call Format (VCF) files.
3. We also compared each sample separately against the reference and therefore a single VCF for each sample was generated

The same conclusions are drawn whether we consider pairwise VCF comparisons or the four-way VCF comparison. The advantage of joining samples in a single four-way VCF was the ease of performing analysis of the samples based on exactly the same positions, and we focus on results from this comparison

8) The largest number of homozygous SNPs was observed on Chromosome 1. Is there an explanation for this observation or is it possible that this is an artifact of the mapping? Chromosome 1 or Chromosome 11 might be the first sequences in the file. If many reads would have been misplaced on these sequences, high numbers of erroneous variants could be expected.

The largest number of homozygous SNPs across the entire genome for all samples was found on chromosomes 4, 5, 11 and 12, not on chromosome 1. We have verified chromosome assignments based on mapped SNP markers from the Array. Chromosome 1 is the largest in physical size with an estimated 108.0 Mb and most assemblies capturing  ~90 Mb. In contrast, chromosome 11 is estimated to be 64.7 Mb with assemblies capturing >50 Mb. Chromosome 6 is the smallest, estimated at 53.8 Mb and assembled lengths in the 45 Mb range.

9) Chromosomal positions “bottom” and “top” could be replaced by “north” and “south”.

We have made this change, using North (top) and South (bottom) at the first use because of convention in the tomato community. Tomato has a long history of cytogenetic research and genetic maps have long been oriented with respect to cytogenetic representations using the top to bottom notation. After this first use we use the North/South notation throughout. Our compromise will help this notation to be understood across research communities and is a useful clarification.

10) The start of the discussion section (multiple paragraphs) is redundant with the results/method section. The authors might want to write it in a more concise way.
We have removed several paragraphs from the discussion to reduce redundancy. We moved a paragraph from the introduction to the discussion as the work from Corn, Soy, and Arabidopsis provides context.

11) The differences between the accessions should be visualized e.g. using a phylogenetic tree. It might be helpful to include an additional data set as some kind of outgroup. Especially to demonstrate high fidelity propagation, a comparison to a very different cultivar is necessary.

We added a new figure as Figure S1. This figure is a cluster dendrogram based on SNPs extracted from the SolCAP data set. We use the accessions ʻHeinz 1706ʼ (gold-standard reference) and S. pimpinellifollium LA1589 as outgroups for the analysis.

Minor comments:
Title: The authors might want to check if two spaces could be removed from the title.
Line 21: The authors might want to rephrase this sentence (“between variation between”). Rephrased.
Line 54: contain > contains ? The reviewer is correct. This paragraph has been moved to the discussion and the correction made.
Line 117: Which function? Fastq-dump? Yes, we added the information.
Line 215: sequence-based SNPs > SNPs. Change made.
Line 219 / 220: The authors might want to rephrase the sentence. We have rephrase this and divided into two sentences.
Parameters of the PCR duplicate removal could be added. Parameters are in the Material and Methods, section 2.4. Sequence alignment and variant calling.  
Line x (line numbers missing): genome > genome sequence We added the missing line numbers
Line x: “artificially condensed” needs to be rephrased. Not the DNA is condensed, but the reference sequence. We rephrased this as “A graphical inspection of the depth of coverage from all genotypes revealed regions in common with extremely high concentration of reads (Figure 1A), probably caused by repetitive DNA or gene-families which have been artificially condensed in the reference.”
Line x: genome reference > reference genome sequence. This sentence was removed from the manuscript.

Submission Date

11 March 2020

Date of this review

14 Mar 2020 07:22:15

Author Response File: Author Response.docx

Reviewer 2 Report

This manuscript shows interesting data. Nevertheless, there are some structural issues that should be addressed in order to take a properly informed and adequate decision.

Title

I suggest that the authors provide a change the title explain precisely the topic of the study and attract the reader's attention.

Introduction

I suggest reviewing, the organization and structure of the introduction. The first part of introduction is focused too many on soybean and maize.It would be nice if the authors provide information regarding: the origin and domestication of Solanum lycopersycum L, analyses of SNP genotyping and re-sequencing in general and specifically applied in tomato (studies regarding comparative transcriptome, and resequencing of wild and cultivated tomatoes), expanding the text with latest references.

Methods
In this section, please provide to divide the chapter into Plant Material (should be revised carefully, as some of the ideas and arguments are not very clear and should be explained more thoroughly) and Genotyping.and data set merging. Please in chapter 2.2 provide to insert a table to clarify the summary of sequencing data and statistics obtained from mapping against the tomato reference genome.

Line 12 Please change “repositories” with “database”

Line 14 Please change with “In order to assess the cultivars “UC82” and “M82” propagation rate and the genetic kinship within them and between germoplasm resources, we used different high-density genotyping and whole genome re-sequencing data of Solanum lycopersicum multiple accessions.”

Line 30 Please change “cultivars” with “accesions”

Line 32 Please change with “In general, the differences within cultivar accessions amount to 0,26% …”

Line 36 Please change with “The data also suggest that these tomato genetic resources propagated with high fidelity.”

Line 68 Please change “Synonyms” with “or”

Line 69 For this sentence “M82 is used as a parent for stable introgression line populations [4,5] serving as a resource for the discovery of quantitative trait loci (QTL) [6–9], map-based cloning [10,11], and genomic studies seeking to link form and function [12,13]” please check writing.

Line 76 Please change “ release” with “released”

Line 80 For this sentence “In an analysis of population structure based on SNP data, ʻM82ʼ and ʻUC 82ʼ clustered together with varieties having pedigrees tracing to California breeding programs “please check writing.

Line 84 Please add comma after accessions

Line 93 Please add Solanaceae Coordinated Agricultural Project (SolCAP)

Line 95 Please explain because it has not been used for comparison. one of S. pimpinellifolium and why used Heinz 1706

Line 107 Please change “downloaded” with “collected”

Author Response

Dear Ms. Rita Ren:

We appreciate the reviewers comments and believe that their questions and suggestions have strengthened the MS. We have addressed comments from the three reviewers (two sent to us via e-mail, and a third which we retrieved from the MDPI portal). To address comments about organization, we have altered the title to provide a less technical and more generally interesting one. We rewrote the abstract to be less technical and reorganized the introduction as well. In order to address the more substantive comments concerning content, we added to a supplemental figure and have now made that figure 1 in the MS. We also added a tree based on clustering to the supplemental files to address a specific suggestion by one of the reviewers. Details and clarification to materials and methods has been in response to specific queries from reviewers. A detail response to the reviewers comments in detail follows.

Thank you for the opportunity to respond.

Yours,

David Francis

Reviewer 2

Open Review

( ) I would not like to sign my review report
(x) I would like to sign my review report

English language and style

( ) Extensive editing of English language and style required
(x) Moderate English changes required
( ) English language and style are fine/minor spell check required
( ) I don't feel qualified to judge about the English language and style

Comments and Suggestions for Authors

This manuscript shows interesting data. Nevertheless, there are some structural issues that should be addressed in order to take a properly informed and adequate decision.

 Title

I suggest that the authors provide a change the title explain precisely the topic of the study and attract the reader's attention. New title: “Propagation Fidelity and Kinship of Tomato Varieties ʻUC 82ʼ and ʻM82ʼ Revealed by Analysis of Sequence Variation”.

 Introduction

I suggest reviewing, the organization and structure of the introduction. The first part of introduction is focused too many on soybean and maize.It would be nice if the authors provide information regarding: the origin and domestication of Solanum lycopersycum L, analyses of SNP genotyping and re-sequencing in general and specifically applied in tomato (studies regarding comparative transcriptome, and resequencing of wild and cultivated tomatoes), expanding the text with latest references.

We have addressed this suggestion by adding a new paragraph to the introduction focus on Solanum lycopersicum resources and we moved information about Soybean and Maize to the discussion.

Methods
In this section, please provide to divide the chapter into Plant Material (should be revised carefully, as some of the ideas and arguments are not very clear and should be explained more thoroughly) and Genotyping.and data set merging. Please in chapter 2.2 provide to insert a table to clarify the summary of sequencing data and statistics obtained from mapping against the tomato reference genome.

Plant material was separated from SNP genotyping and whole genome re-sequencing data. Table S1 was added in order to clarify the source of the samples used. Summary of sequencing data and statistics obtained from mapping against the tomato reference genome is presented in Table 2.

 Line 12 Please change “repositories” with “database” We have made this change.

Line 14 Please change with “In order to assess the cultivars “UC82” and “M82” propagation rate and the genetic kinship within them and between germoplasm resources, we used different high-density genotyping and whole genome re-sequencing data of Solanum lycopersicum multiple accessions.” We have made this change.

Line 30 Please change “cultivars” with “accesions”  We have made this change.

Line 32 Please change with “In general, the differences within cultivar accessions amount to 0,26% …” We have made this change: “In general, the differences within variety accessions amount to 0,26% …”.

Line 36 Please change with “The data also suggest that these tomato genetic resources propagated with high fidelity.” We have made this change.

Line 68 Please change “Synonyms” with “or” We have change this to “also referred to as”.

Line 69 For this sentence “M82 is used as a parent for stable introgression line populations [4,5] serving as a resource for the discovery of quantitative trait loci (QTL) [6–9], map-based cloning [10,11], and genomic studies seeking to link form and function [12,13]” please check writing. Changed to “M82 is used as a parent for stable introgression line populations [4,5] which serve as a resource for the discovery of quantitative trait loci (QTL) [6–9], map-based cloning [10,11], and genomic studies seeking to link form and function [12,13]”

Line 76 Please change “ release” with “released”. We believe that “release” is the correct use, here. To help with clarity we changed this to read “variety release notification”.

Line 80 For this sentence “In an analysis of population structure based on SNP data, ʻM82ʼ and ʻUC 82ʼ clustered together with varieties having pedigrees tracing to California breeding programs “please check writing. We believe this syntax is correct.

Line 84 Please add comma after accessions. We have made this change.

Line 93 Please add Solanaceae Coordinated Agricultural Project (SolCAP). We have made this change.

Line 95 Please explain because it has not been used for comparison. one of S. pimpinellifolium and why used Heinz 1706. We have explained that Heinz 1706 is the gold-standard reference.

Line 107 Please change “downloaded” with “collected”. We believe that downloaded is more accurate.

Submission Date

11 March 2020

Date of this review

26 Mar 2020 15:53:45

Author Response File: Author Response.docx

Reviewer 3 Report

Cambiaso and his colleagues accessed high-density genotyping and resequencing data for multiple accessions of two tomato cultivars including UC 82 and M82. The results revealed that the differences were 0.26% for genotyping data and up to 0.67% for sequencing data within cultivar accessions. This manuscript looks organized well and easy to follow. I still have couple of concerns need to be addressed.

• In table 1, on chromosome 6, there is 1 bp for polymorphic region. It looks the data is confusing for me. Can I know the criteria you were looking for polymorphic SNPs?
• Table 3, the first and second columns looks mismatch. It would be better to fix it.
• For abstract, it would be better if you would highlight the key point and shorten it.

Author Response

Dear Ms. Rita Ren:

We appreciate the reviewers comments and believe that their questions and suggestions have strengthened the MS. We have addressed comments from the three reviewers (two sent to us via e-mail, and a third which we retrieved from the MDPI portal). To address comments about organization, we have altered the title to provide a less technical and more generally interesting one. We rewrote the abstract to be less technical and reorganized the introduction as well. In order to address the more substantive comments concerning content, we added to a supplemental figure and have now made that figure 1 in the MS. We also added a tree based on clustering to the supplemental files to address a specific suggestion by one of the reviewers. Details and clarification to materials and methods has been in response to specific queries from reviewers. A detail response to the reviewers comments in detail follows.

Thank you for the opportunity to respond.

Yours,

David Francis

Reviewer 3

Open Review

(x) I would not like to sign my review report 
( ) I would like to sign my review report 

English language and style

( ) Extensive editing of English language and style required 
(x) Moderate English changes required 
( ) English language and style are fine/minor spell check required 
( ) I don't feel qualified to judge about the English language and style 

Comments and Suggestions for Authors

Cambiaso and his colleagues accessed high-density genotyping and resequencing data for multiple accessions of two tomato cultivars including UC 82 and M82. The results revealed that the differences were 0.26% for genotyping data and up to 0.67% for sequencing data within cultivar accessions. This manuscript looks organized well and easy to follow. I still have couple of concerns need to be addressed.

• In table 1, on chromosome 6, there is 1 bp for polymorphic region. It looks the data is confusing for me. Can I know the criteria you were looking for polymorphic SNPs? Chromosome 6 had a single SNP and therefore it was difficult to define a “region” as we did for other chromosomes. We have added a footnote in explanation.
• Table 3, the first and second columns looks mismatch. It would be better to fix it. We have made this change.

• For abstract, it would be better if you would highlight the key point and shorten it. We have made this change.

Submission Date

11 March 2020

Date of this review

30 Mar 2020 05:33:05

Author Response File: Author Response.docx

Round 2

Reviewer 1 Report

Major comments:

1) ok

2) Substantial improvement. The abstract is well written and easy to understand. It is not necessary to mention that the variants are stored in a VCF file as this is a minor technical detail.

3) ok

4) ok

5) ok

6) ok

7) ok

8) ok

9) If north/south is not common in the tomato community, the authors do not have to adopt it.

10) ok

11) Figure S1: The authors might want to change the y-axis label.

Minor comments:

SNP / SNV: The authors should use one term consistently throughout the manuscript.

genome / genome sequence: The authors should carefully check if some occurrences of "genome" should be changed to "genome sequence".

sequence-based SNPs > SNPs

supplemental file Table S2 > Table S2

Author Response

Dear Ms. Rita Ren:

We appreciate the suggestions in this second round of revisions and believe that the detailed reviewer comments have strengthened the MS. We have addressed all the reviewer comments and those of the editor. Thank you for the opportunity to respond.

Yours, David Francis

Editor Comments

Instruction for authors:

Authority on Latin binomial should be provided after each common name the first time referred to in the text.

We use Solanum lycopersicum and Solanum pimpinellifolium. For both the Authority is Linnaeus. We have added “L.” at only the first use in the text of the MS.

Number of significant digits should be based on the precision of the analytical method and round accordingly and presented for a variable should be correct and consistent.

We have corrected percentage values to numbers of significant digits from Table 1, 2 and 3. We also corrected mean depth of coverage in Table 2.

Conclusions: This section is mandatory, should be added to the manuscript.

There is a conclusion section (5) at line 1490.

Authors are encouraged to provide a *graphical abstract* or *video* to present their research.

We do not feel that a graphical abstract or video would be an effective way to present this work, Figure 2 and Figure S1 both summarize the work in graphical form.

Open Review

(x) I would not like to sign my review report 
( ) I would like to sign my review report 

English language and style

( ) Extensive editing of English language and style required 
( ) Moderate English changes required 
(x) English language and style are fine/minor spell check required 
( ) I don't feel qualified to judge about the English language and style 

Comments and Suggestions for Authors

Major comments:

1) ok

2) Substantial improvement. The abstract is well written and easy to understand. It is not necessary to mention that the variants are stored in a VCF file as this is a minor technical detail. We have made this change

3) ok

4) ok

5) ok

6) ok

7) ok

8) ok

9) If north/south is not common in the tomato community, the authors do not have to adopt it. We reverted to the nomenclature that is convention in the tomato community.

10) ok

11) Figure S1: The authors might want to change the y-axis label. We have changed the y-axis label from “Height” to “Distance”.

Minor comments:

SNP / SNV: The authors should use one term consistently throughout the manuscript. We have replaced “variant” with SNP in most cases. For example: headers 2.4 and 3.2, Figure 1 and Table 3. We did not replace it in Table 2 because Indels and SNPs are shown.

genome / genome sequence: The authors should carefully check if some occurrences of "genome" should be changed to "genome sequence". We have carefully checked all the occurrences of “genome” and changed to “genome sequence” when it was more accurate.

sequence-based SNPs > SNPs We have made this change.

supplemental file Table S2 > Table S2 We have made this change.

Submission Date

11 March 2020

Date of this review

03 Apr 2020 13:40:37

Author Response File: Author Response.docx