The Use of DArTseq Technology to Identify Markers Related to the Heterosis Effects in Selected Traits in Maize
Round 1
Reviewer 1 Report
The Use of DArTseq Technology to Identify Markers Related to the Heterosis Effects in Selected Traits in Maize
Abstract
Short the abstract and improve it.
Materials and Methods
Principal component analysis: explain this
give plan of work in figures or in simple model format.
Explain sowing method, date and time of both localities.
Results
First explain PC analysis.
Explain 1st year data
2nd year data
combine effect of both years
Add some of graphs, dendrograms, figures etc.
How about the years 12, 13,14????
You are comparing with a particular hybrid, did mention anywhere, either increasing decreasing what is happening??
How about the DarTseq comparison of particular hybrid or reference, in terms of molecular.
what about the localities
what is the difference in two or four localities/environment?
Explain all the data together, give graphs and figures etc.
Overall improve the writeup and English.
Author Response
Response to Reviewer 1 Comments
Reviewer #1
Point 1: Abstract
Short the abstract and improve it..
Response: We have shortened and improved Abstract. New abstract: “Spectacular scientific advances in the area of molecular biology and the development of modern biotechnological tools have had a significant impact on the development of maize heterosis breeding. One technology based on next-generation sequencing is DArTseq. The plant material used for the research consisted of 13 hybrids resulting from crossing of inbred maize lines. A two-year field experiment was established at two Polish breeding stations: Smolice and Łagiewniki. Nine quantitative traits were observed: cob length, cob diameter, core length, core diameter, number of rows of grain, number of grains in a row, mass of grain from the cob, weight of one thousand grains, and yield. The isolated DNA was subjected to DArTseq genotyping. Association mapping was performed using a method based on the mixed linear model. A total of 81602 molecular markers (28571 SNPs and 53031 SilicoDArTs) were obtained as a result of next-generation sequencing. Out of 81602, 15409 (13850 SNPs and 1559 SilicoDArTs) were selected for association analysis. The 105 molecular markers (eight SNPs and 97 SilicoDArTs) were associated with the heterosis effect of at least one trait in at least one environment. A total of 186 effects were observed. The number of statistically significant relationships between molecular marker and heterosis effect varied from eight (for cob length) and nine (for yield) to 42 (for the number of rows of grain). Of particular note are three markers (2490222, 2548691 and 7058267), which were significant in 17, 8 and 6 cases, respectively. Two of them (2490222, 7058267) were associated with heterosis effects of yield in three of the four environments.”.
Materials and Methods
Point 2: Principal component analysis: explain this.
Response: We have shortened the description of the methodology. We did not use standard principal component analysis in this paper. We only used eigenvalues from PCA to assess population structure with a mixed model. This approach is standard for association mapping.
Point 3: give plan of work in figures or in simple model format.
Response: We presented plan of work in the Figure 1.
Point 4: Explain sowing method, date and time of both localities.
Response: We added information in the text of the manuskript: “Maize was sown in each year of the study on April 20. Mineral fertilization was adapted to the needs of maize grown for grain harvest. 350 kg ha–1 of polyphoska and 160 kg ha–1 of urea were applied. The nutrient content was: before winter ploughing: N – 21 kg N ha–1 in the form of polyphoska, P – 70 kg P2O5 ha-1 in the form of polyphoska, K – 105 kg K2O ha–1 in the form of polyphoska. Before sowing, N – 73.6 kg N ha–1 in the form of urea was applied. Chemical control of weeds was done with a field sprayer.”.
Results
Point 5: First explain PC analysis.
Response: We did not use standard principal component analysis in this paper. We only used eigenvalues from PCA to assess population structure with a mixed model. This approach is standard for association mapping.
Point 6: Explain 1st year data. 2nd year data.
Response: Due to statistically significant interactions: genotypes-years, genotypes-locations, years-locations, and genotypes-years-locations, association analyses were performed independently for four environments (combinations of years-locations). We have added this information in the text of the manuscript.
Point 7: combine effect of both years.
Response: The statistical significance of the interactions years-locations and genotypes-years-locations excludes the possibility of analyzing for both years simultaneously. Such an approach would be methodologically incorrect, and the results obtained would be biased by ignoring significant interactions.
Point 8: Add some of graphs, dendrograms, figures etc.
Response: We have added Figures showing: the values of heterosis effects; classified by the studied environments (box-and-whisker diagrams) and e number of statistically significant relationships between molecular marker and heterosis effect for particular traits in studied environments.
Point 9: How about the years 12, 13,14????
Response: We are very sorry for the mistake that was made. The experiment was conducted in 2013 and 2014, and we mistakenly put '2012'. We have corrected our earlier oversight. Thank you very much for catching our mistake.
Point 10: You are comparing with a particular hybrid, did mention anywhere, either increasing decreasing what is happening??
Response: The main objective of our work was to propose a method for detecting markers coupled to the effects of heterosis of quantitative traits. The results obtained show the versatility of the given markers as tools for heterosis diagnosis. The advantage of the proposed approach is that it is independent of the sign of the heterosis effect. We supplemented the Conclusions section with this proposal.
Point 11: How about the DarTseq comparison of particular hybrid or reference, in terms of molecular.
Response: We added text: “Genotyping using DArTseq technology was performed by Diversity Arrays Technology Pty Ltd. (Kirian A., Canberra, Australia). DNA sample digestion/ligation reactions were processed according to Kilian and Graner [20], but replacing a single PstI–compatible adaptor with two adaptors corresponding to PstIand NspI–compatible sequences and moving the assay on the sequencing platform as described by Sansaloni et al. [24]. The PstI–compatible adapter was designed to include Illumina flowcell attachment sequence, sequencing primer sequence, and “staggered” varying length barcode region, similar to the sequence reported by Elshire et al. [25]. Reverse adapter contained flowcell attachment region and NspI–compatible overhang sequence. Only “mixed fragments” (PstI–NspI) were amplified in PCR using the following reaction conditions: denaturation, 1 min at 94°C; followed by 30 cycles of 94°C for 20 s, 57°C for 30 s, and 72°C for 45 s; and the final elongation, 72°C for 7 min. After PCR, equimolar amounts of amplification products from each sample of the 96–well microtiter plate were bulked and applied to c–Bot (Illumina) bridge PCR, followed by sequencing on Illumina Hiseq2500. The sequencing (single read) was run for 78 cycles. Sequences generated from each lane were processed using proprietary DArT analytical pipelines. In the primary pipeline, the fastq files were first processed to filter away poor-quality sequences, applying more stringent selection criteria to the barcode region compared to the rest of the sequence. In that way, the assignments of the sequences to specific samples carried in the “barcode split” step were very reliable. Approximately 2500000 (±7%) sequences per barcode/sample were used in marker calling. Finally, identical sequences were collapsed into “fastqcall files”. These files were used in the secondary pipeline for DArT PL’s proprietary SNP and SilicoDArT (presence/absence of restriction fragments in representation) calling algorithms (DArTsoft14).”.
Point 12: what about the localities.
Response: We added text: “Łagiewniki and Smolice are 137 km apart. Both locations were chosen because they differ in soil type and rainfall. In Łagiewniki, the soil quality classification is predominantly class I to III, while in Smolice it is class III to V. The average annual rainfall in Łagiewniki is 560-660 mm while in Smolice it is 390-460 mm. The locations, which vary in terms of environmental conditions, were chosen to test the influence of the environment on the magnitude of the heterosis effect.”.
Point 13: what is the difference in two or four localities/environment?
Response: Differences in the expression (location and effect values) of genes associated with heterosis effects are primarily due to genotype-environment interactions, which are partly influenced by weather conditions.
Point 14: Explain all the data together, give graphs and figures etc. .
Response: We have added Figures showing: the values of heterosis effects; classified by the studied environments (box-and-whisker diagrams) and e number of statistically significant relationships between molecular marker and heterosis effect for particular traits in studied environments.
Point 15: Overall improve the writeup and English.
Response: We have corrected the English throughout the manuscript.
Author Response File: 
Author Response.pdf
Reviewer 2 Report
The manuscript deals with the use of DArTseq technology to identify markers related to the heterosis effects in maize. More specifically, the authors focus on the detection of SNPs and DArTseq markers for cob length, cob diameter, core length, core diameter, the number of rows of grain, the number of grains in a row, the mass of grains per cob, weight of one thousand grains, and yield. It is written in a good scientific manner and proper English (including scientific terms) is used throughout the text. Unfortunately, in spite of the large number of molecular markers obtained (total 81602, of which 28571 SNPs and 53031 SilicoDArTs), relatively few markers were associated with the heterosis effect of at least one trait. Most disappointing is that the heterosis effect for the most complex (and most interesting) trait - yield - was determined by just five markers (of which only 1 SNP). This, however, is not a shortcoming of the methodology as similar numbers were obtained in other studies using the same approach (properly referenced by the authors). While demonstrating the usefulness of the selected approach this also very well demonstrates its limitations. As the authors properly note in the Discussion (Lines 373-374), the usefulness of this method comes into play in cases when the criterion of time is more important than the initial financial burden. This part of the discussion is one of the main contributions of the manuscript that often lacks from other papers dealing with whole genome sequencing (or other high-volume sequencing approaches)..
Author Response
Response to Reviewer 2 Comments
Reviewer #2
Point 1: The manuscript deals with the use of DArTseq technology to identify markers related to the heterosis effects in maize. More specifically, the authors focus on the detection of SNPs and DArTseq markers for cob length, cob diameter, core length, core diameter, the number of rows of grain, the number of grains in a row, the mass of grains per cob, weight of one thousand grains, and yield. It is written in a good scientific manner and proper English (including scientific terms) is used throughout the text. Unfortunately, in spite of the large number of molecular markers obtained (total 81602, of which 28571 SNPs and 53031 SilicoDArTs), relatively few markers were associated with the heterosis effect of at least one trait. Most disappointing is that the heterosis effect for the most complex (and most interesting) trait - yield - was determined by just five markers (of which only 1 SNP). This, however, is not a shortcoming of the methodology as similar numbers were obtained in other studies using the same approach (properly referenced by the authors). While demonstrating the usefulness of the selected approach this also very well demonstrates its limitations. As the authors properly note in the Discussion (Lines 373-374), the usefulness of this method comes into play in cases when the criterion of time is more important than the initial financial burden. This part of the discussion is one of the main contributions of the manuscript that often lacks from other papers dealing with whole genome sequencing (or other high-volume sequencing approaches)..
Response: Thank you very much. We are very grateful for the appreciation of our research.
Author Response File: Author Response.pdf
Round 2
Reviewer 1 Report
-Good effort
Materials and methods
2.3 Genotyping as well as SNP and SilicoDArT Data Processing
Add, sampling method.
How many samples, DNA sample digestion/ligation reactions, how you take samples before/after harvesting from which location you take samples or its simple field experiment along with molecular studies.
How many samples you sequenced?
or you take samples from best performing phenotypes like high yielding line or lines.
Where is sequenced data???? either online or supplementary files etc.
any pics/graph of PCR/ gel etc.
Please add as many things you have, it is very informative research, you have worked a lot but added a little.
Improve English, in new writeup.
Accept all comments/add to manuscript.
Author Response
Response to Reviewer 1 Comments
Reviewer #1
Point 1: Good effort.
Response: Thank you very much.
Materials and methods
2.3 Genotyping as well as SNP and SilicoDArT Data Processing
Point 2: Add, sampling method.
Response: The manuscript was completed with the following text.
“Seeds of the F1 generation hybrids (10 from each hybrid) were sown onto sterile Pettri plates to obtain young seedlings. Approximately 1 cm2 of leaf tissue was taken from each of the 10 plants and a collection sample was prepared (separately for each hybrid). The plant material prepared in this way was isolated. Isolation was carried out using Maxwell equipment from Promega, for automatic DNA isolation. The Maxwell® RSC Tissue DNA Kit AS1610 was used for this purpose. After isolation, the concentration and purity of the DNA samples were tested and equal dilutions were prepared to 100 ng µl–1. DNA concentration and purity were checked using a DeNovix spectrophotometer from TK Biotech. Care was taken to select only samples for sequencing where the DNA purity at 260/280 absorbance was between 1.8 and 2.0 and at 230/260 absorbance was not less than 1.8. DNA prepared in this way was placed in a 96-well plate and sent for sequencing, to Diversity Arrays Technology.”
Point 3: How many samples, DNA sample digestion/ligation reactions, how you take samples before/after harvesting from which location you take samples or its simple field experiment along with molecular studies.
Response: The manuscript was completed with the following text. “The seeds of each hybrid were divided. One part was allocated to set up a field experiment and 10 pieces were allocated for molecular analyses.” The method of collecting material for isolation is shown above.
Point 4: How many samples you sequenced? or you take samples from best performing phenotypes like high yielding line or lines.
Response: The manuscript was completed with the following text. “For next-generation sequencing, F1 hybrids and inbred lines (highly homozygous), which were the parental forms of these hybrids, were sent to a 96-well plate. However, for the above publication, only the sequencing results of the F1 hybrids were used from the database of results obtained. The results for the lines were used for a different inference. The inbred lines were obviously characterized by low yield because they had previously undergone inbreeding to obtain plant materials that were highly homozygous. The F1 hybrids were characterized by high yield.”
Point 5: Where is sequenced data???? either online or supplementary files etc.
Response: Sequence data, due to its very large capacity, can be found (too large for supplementary material) at the correspondent author and at https://www.diversityarrays.com.
Point 6: any pics/graph of PCR/ gel etc.
Response: The study did not require PCR analyses to check the quality of the DNA because the quality was checked on a spectrophotometer. These were the guidelines on the Diversity Arrays Technology website for preparing DNA for sequencing.
Point 7: Please add as many things you have, it is very informative research, you have worked a lot but added a little.
Response: Thank you very much for your suggestion. We have added as many results as possible regarding sampling, sequencing and association mapping results of heterosis effect of selected maize traits. We are aware that due to the pioneering approach to this issue, the information provided is a very important part of the manuscript.
Point 8: Improve English, in new writeup.
Response: English has been improved.
Point 9: Accept all comments/add to manuscript.
Response: The manuscript was completed.
Author Response File: Author Response.pdf
