Next Generation Mapping of Enological Traits in ... - Semantic Scholar

RESEARCH ARTICLE

Next Generation Mapping of Enological Traits in an F2 Interspecific Grapevine Hybrid Family Shanshan Yang1☯, Jonathan Fresnedo-Ramírez2☯, Qi Sun2, David C. Manns3, Gavin L. Sacks4, Anna Katharine Mansfield3, James J. Luby5, Jason P. Londo6, Bruce I. Reisch1, Lance E. Cadle-Davidson6, Anne Y. Fennell7,8* 1 Horticulture Section, School of Integrative Plant Science, Cornell University, Geneva, New York, United States of America, 2 Bioinformatics Facility, Institute of Biotechnology, Cornell University, Rhodes Hall, Ithaca, New York, United States of America, 3 Department of Food Science, Cornell University—NYSAES, Geneva, New York, United States of America, 4 Department of Food Science, Cornell University, Ithaca, New York, United States of America, 5 Department of Horticultural Science, University of Minnesota, St. Paul, Minnesota, United States of America, 6 USDA-ARS Grape Genetics Research Unit, Geneva, New York, United States of America, 7 Plant Science Department, South Dakota State University, Brookings, South Dakota, United States of America, 8 BioSNTR, Brookings, South Dakota, United States of America

OPEN ACCESS Citation: Yang S, Fresnedo-Ramírez J, Sun Q, Manns DC, Sacks GL, Mansfield AK, et al. (2016) Next Generation Mapping of Enological Traits in an F2 Interspecific Grapevine Hybrid Family. PLoS ONE 11(3): e0149560. doi:10.1371/journal.pone.0149560 Editor: Yuepeng Han, Wuhan Botanical Garden of Chinese Academy of Sciences, CHINA Received: December 19, 2015 Accepted: February 2, 2016 Published: March 14, 2016 Copyright: This is an open access article, free of all copyright, and may be freely reproduced, distributed, transmitted, modified, built upon, or otherwise used by anyone for any lawful purpose. The work is made available under the Creative Commons CC0 public domain dedication. Data Availability Statement: All relevant data are within the paper and its Supporting Information files. All sequence information for the sequence markers are published in S7 File. Funding: This work was supported by the United States Department of Agriculture, National Institute of Food and Agriculture Specialty Crops Research Initiative Award No. 2011-51181-30635, BIR, LCD, AYF, GLS, JJL, AKM. This fund supported JFR and DCM. This work was also supported by the National Science Foundation Award No. IIA1355423, South Dakota Research Innovation Center and a gift from J. Lohr Vineyards & Wines to AYF. SY was supported

☯ These authors contributed equally to this work. * [email protected]

Abstract In winegrapes (Vitis spp.), fruit quality traits such as berry color, total soluble solids content (SS), malic acid content (MA), and yeast assimilable nitrogen (YAN) affect fermentation or wine quality, and are important traits in selecting new hybrid winegrape cultivars. Given the high genetic diversity and heterozygosity of Vitis species and their tendency to exhibit inbreeding depression, linkage map construction and quantitative trait locus (QTL) mapping has relied on F1 families with the use of simple sequence repeat (SSR) and other markers. This study presents the construction of a genetic map by single nucleotide polymorphisms identified through genotypingby-sequencing (GBS) technology in an F2 mapping family of 424 progeny derived from a cross between the wild species V. riparia Michx. and the interspecific hybrid winegrape cultivar, ‘Seyval’. The resulting map has 1449 markers spanning 2424 cM in genetic length across 19 linkage groups, covering 95% of the genome with an average distance between markers of 1.67 cM. Compared to an SSR map previously developed for this F2 family, these results represent an improved map covering a greater portion of the genome with higher marker density. The accuracy of the map was validated using the well-studied trait berry color. QTL affecting YAN, MA and SS related traits were detected. A joint MA and SS QTL spans a region with candidate genes involved in the malate metabolism pathway. We present an analytical pipeline for calling intercross GBS markers and a high-density linkage map for a large F2 family of the highly heterozygous Vitis genus. This study serves as a model for further genetic investigations of the molecular basis of additional unique characters of North American hybrid wine cultivars and to enhance the breeding process by marker-assisted selection. The GBS protocols for identifying intercross markers developed in this study can be adapted for other heterozygous species.

PLOS ONE | DOI:10.1371/journal.pone.0149560 March 14, 2016

1 / 19

Mapping an F2 Grapevine Population

by a grant from the National Grape and Wine Initiative to BIR. Any opinions, findings, and conclusions or recommendations expressed in this publication are those of the authors and do not necessarily reflect the views of the NSF, USDA or other funding entities. The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript. Competing Interests: SY was supported by a grant from the National Grape and Wine Initiative to BIR. This does not alter the authors' adherence to PLOS ONE policies on sharing data and materials. The specific roles of these authors are articulated in the ‘author contributions’ section.

Introduction The inherent role of the enological properties of a winegrape cultivar is undeniable in the final taste of wine. While both growing conditions and choice of vinification procedures influence eventual wine properties [1–4], the grapevine cultivar has arguably the largest effect on wine quality [5, 6]. Understanding the genetic controls of enologically important traits will assist breeders, viticulturists, and enologists to delineate strategies for improving wine quality. Grapevine is a cross pollinating, highly heterozygous crop that exhibits inbreeding depression in the form of diminished seed viability and reduced vine vigor. Because of this, grapevine breeders typically employ breeding schemes based on background selection for V. vinifera fruit quality and foreground selection for introgressed traits [7, 8], which are applied under a framework of modified-backcrossing [2, 8]. Winegrape breeding began in earnest in the 19th C., as breeders attempted to develop grapevines with improved pest and disease resistance by cross-hybridizing popular V. vinifera winegrape cultivars with resistant North American wild species such as V. aestivalis Michx, V. cinerea (Engelm. ex A. Gray) Engelm. ex Millard, V. labrusca L., V. riparia Michx and V. rupestris Scheele [9].While these introgressions have positive adaptive traits including disease resistance, pest resistance, and cold hardiness [5, 10, 11], they often have inferior fruit composition for winemaking [9, 12]. Common quality parameters for wine grapes include titratable acidity (TA), soluble solids (SS) content as a proxy for sugar concentration, and yeast assimilable nitrogen (YAN) [13]. Sugars increase during berry ripening, and an SS of 20–24% w/w (Brix) is typically targeted for production of a dry table wine with 11–14% v/v alcohol. Conversely, malic acid (MA)–one of the two major grape organic acids that contribute to TA–is respired during berry ripening [14]. While SS contents of wild Vitis and V. vinifera are comparable, MA concentrations in wild Vitis are generally much higher–often in excess of 10 g/L at harvest [15]–and wines produced from Vitis interspecific hybrids are frequently reported to have excessive acidity and sourness [5] if not ameliorated before bottling. This situation is exacerbated in cooler climates where hybrid grapes are more widely used, such as the northern United States, because of slower MA respiration [16]. Thus, a key goal following trait introgression is combining lower MA content with appropriate sugar content. YAN is also widely measured in winegrapes prior to fermentation, since insufficient concentrations (< 150 mg/L) can lead to stuck fermentations and offodor formation [17], while excessive concentrations can lead to formation of ethyl carbamate, a known carcinogen [1]. YAN concentrations at both extremes have been reported in interspecific hybrids [18], and breeding cultivars with appropriate YAN content is desirable. Genetic controls of these key compositional factors have only recently been investigated in grapevine [19]. The formation of MA in the grape berry involves the glycolysis and citric acid (TCA) cycle and is regulated by the malate metabolism pathway. The diverse pathways in metabolic processes and environmental factors, such as temperature during ripening, all contribute to the complexity for dissecting the molecular basis of MA content. Two recent studies reported several QTLs controlling total SS and acid content [20, 21]. However, the identified QTLs could only explain a small percentage of the total phenotypic variance (< 17%), and candidate genes could not be identified. QTL analysis of YAN in grape berries has not been attempted, even though YAN is known to vary considerably among both V. vinifera [22] and Vitis spp. hybrids [18]. Given the heterozygosity and inbreeding depression in Vitis, QTL mapping studies generally use F1 full-sib progenies and the pseudo-testcross method [23], with 70 to 300 progeny and from 100 up to 1,826 molecular markers [20, 24–27]. Previously, a family of 119 F2 progeny was developed from selfing an F1 individual derived from the cross V. riparia × Vitis hybrid


2 / 19


‘Seyval’ [28], and a linkage map based on 120 simple sequence repeats (SSRs) was developed [29]. ‘Seyval’ is a complex interspecific hybrid of V. vinifera (55% by pedigree) with wild species native to the south-central United States and is a relatively cold hardy white wine cultivar, resistant to disease and the phylloxera aphid Daktulosphaira vitifoliae (Fitch 1855). V. riparia has been used to breed for early acclimation and freezing tolerance in hybrid winegrapes [28]. The acclimation and dormancy characteristics of V. riparia and ‘Seyval’ have been studied at the physiological, transcriptomic and metabolomic levels providing additional gene level characterization [28–32]. Genotyping by sequencing (GBS) is a marker platform that uses restriction enzymes to reduce genome complexity prior to next generation sequencing. GBS has become widely used due to several benefits: simultaneous marker discovery and genotyping, high sample throughput and scalability, and high resolution at a low per-marker cost [33]. However, due to arbitrary sampling of sites and the high level of multiplexing typical in GBS, sequencing depth is usually reduced, leading to missing data and heterozygote undercalling, in which heterozygous sites are undersampled and wrongly interpreted as homozygous [34]. While heterozygote undercalling has been irrelevant to GBS applications in inbred species, it is a critical problem in grapevine and other highly heterozygous and diverse species, requiring the development of new methodologies. The present study demonstrates the use of novel GBS analysis methods to develop an improved linkage map for the F2 mapping family derived from V. riparia × ‘Seyval’ [35]. This GBS map was validated by analysis of the berry skin color locus, previously characterized [36– 38]. Our results indicate this is an effective approach for mapping QTL and identifying candidate genes in Vitis, including preliminary results for SS, MA and YAN.

Materials and Methods Plant material The F2 mapping family was generated as described in Fennell et al. [28]. The F2 progeny were derived from self-pollination of a single hermaphrodite F1 individual (16-9-2) generated from a cross of V. riparia (USDA PI 588259) × ‘Seyval’ (Seyve-Villard 5–276). One hundred nineteen F2 seedlings were germinated from seed, grown in the greenhouse, cycled into dormancy and cuttings taken for vegetative propagation. After chilling fulfillment, cuttings were rooted and vines grown for field establishment as well as maintenance in a controlled environment. Ecodormant one-year-old vines were planted in the field in spring 2005, trained and maintained for fruiting. A first genetic map for this family was published in 2009 using SSR markers [29]. Fruit from the F1 parent, the V. riparia grandparent, and 65 fruiting progeny with greater than 20 clusters were harvested 30 days post veraison, and 150 g of randomly collected berries were flash frozen and sent to Cornell University for berry quality phenotyping in 2013. Fruit ripening parameters particularly SS and acid content are impacted by changes in temperature. This population had ripening times spanning early to late season; therefore, harvesting at 30 days post veraison avoided low night temperature exposure that would have impacted the late season genotypes. Additional F2 seeds were generated in 2011 and 2012. Three hundred fourteen additional seedlings were germinated in 2013 and genotyped for increased map resolution.

Phenotyping Frozen berry samples (150 g) from each individual were destemmed, allowed to partially thaw, and homogenized in a 250-mL stainless steel Waring blender (Waring laboratory science, Stamford, CT, USA) at medium speed for 30 s followed by high speed for 30 s. A 10 g portion


3 / 19


of berry slurry was transferred to a 15-mL plastic centrifuge tube and then frozen at -20°C. Prior to analysis, samples were thawed. Juice soluble solids (Brix) were analyzed by refractometer (Misco Model #PA203X; Misco, Cleveland, OH USA). The MA was measured on an Agilent 1260 infinity series HPLC using a previously reported method [39]. The YAN was measured as the sum of ammonium and primary amino nitrogen using commercial enzymatic assays (Unitech Scientific, LLC., Hawaiian Gardens, CA, USA) on a ChemWell model 2900 analyzer (Awareness Technologies, Palm City, FL, USA). A subset of MA analyses (one out of every ten individuals) were subjected to either injection replication or extraction replication. The relative standard deviation (%RSD) for MA analyses was 0.4% and 0.7% for injection alone or extraction/injection, respectively. A set of standards and blanks was run every ten samples to verify the retention time and response factor of analytes being measured. An analogous approach was used for replication of YAN analyses. Technical replicates of SS measurements were not conducted due to the high reproducibility of the method.

Genotyping For each vine, a single small leaf (less than 1 cm diameter) was placed in a tube of a Costar 96-well cluster tube collection plate (Corning Life Sciences, Tewksbury, MA, USA). Each 96-well plate consisted of 91 genotype-specific samples, two sets of duplicates and one blank well. The location of the blank well in each plate was unique as a plate identity control. Leaf tissues were maintained at 4°C during and after harvest. In the laboratory, two stainless steel genogrinder beads were placed in each tube and plates were frozen at -80°C. Tissue grinding took place in a Geno/Grinder 2000 (OPS Diagnostics LLC, Lebanon NJ, USA) with 96-well plates agitated in pairs at 400X speed for 1 minute. Plates were then stored at -80°C until processing with DNeasy 96-well DNA extraction kits (Qiagen, Valencia CA, USA). Modifications were made to the manufacturer’s protocol to improve DNA quality and quantity as follows: 1) PVP40 (2% w/v) was added to the AP1 lysis buffer prior to heating; and 2) visual inspection for complete re-suspension of the sample pellet of each 8-tube strip was added to the agitation step following AP1 addition. Genotyping-by-sequencing (GBS) was performed as described by Elshire et al. (2011) [35], integrating four 96-well plates across 384 barcodes for library preparation and sequencing. For SNP calling, the raw sequence data for the 424 F2 progeny plus the F1 progenitor (16-9-2) was processed through the TASSEL 3.0 GBS pipeline [33] using the 12X.v2 V. vinifera ‘PN40024’ reference genome [40] from The French-Italian Public Consortium (https://urgi.versailles.inra. fr/Species/Vitis/Data-Sequences/Genome-sequences) for alignment and the Burrows–Wheeler Aligner (BWA) mem [41] with default parameters. The output consisted of variant call format (VCF) file version 4.1 [42] including SNPs present in at least 40% of the progeny and with a minor allele frequency (MAF) 0.1. Subsequently, the VCF was filtered using vcftools ver. 1.12a [42] and TASSEL [43] versions 3.0 and 4.0. A total of 291,453 SNPs were identified in 424 F2 progeny by TASSEL 3.0, then a custom filtering process was applied for alignment. The filtering was based on keeping sites with a minimum read depth of 6 and 75% completeness by site across progeny and by progeny across sites. Results were output as a TASSEL hapmap file. Finally, using a custom perl script (S1 File) markers heterozygous in the F1 progenitor and with a co-dominant 1:2:1 segregation among the F2 progeny were identified by a chi-squared (χ2) goodness-of-fit test at α0.01. These were reformatted to be imported in JoinMap1 4.1.

Map construction For map construction, the intercross markers described above were coded as hk×hk (assumption of no dominance because the gametic phases were unknown) to be imported in


4 / 19


JoinMap41 4.1 Build: 31jul13.4feb11. Cross-pollination (CP) cross type was used to estimate linkages [44]. This type of marker segregation is particularly useful when performing linkage mapping in outbreeding species because the markers are symmetrical with respect to parental segregation and because they allow construction of integrated linkage maps even when the identification of tri or tetra-allelic markers is insufficient [45]. For the present study, these markers, in conjunction with the multipoint maximum likelihood algorithm implemented in JoinMap1 [46], allowed linkage calculation and phasing determination, which enabled us to recode marker genotypes based on predicted phase, and then perform QTL analyses as in an F2 experimental cross. Prior to grouping and ordering, markers with highly significant (p0.00001) deviation from Mendelian expectations and loci duplicating genetic information were filtered in JoinMap1. For grouping, a linkage-independence LOD 10 was used. For ordering and genetic location determination, the Maximum Likelihood mapping algorithm and Kosambi mapping function were applied, in addition to three rounds of map optimization. Given the amount of missing data allowed, genetic distance inflation occurred among markers. Thus, to keep the most reliable markers and their locations while limiting distance inflation, a threshold of 2.5 cM for the nearest neighbor stress (N.N. stress) was considered. This is an empirical threshold determined through the linkage mapping of 16 full-sib families analyzed in the USDA-NIFA Specialty Crops Research Initiative VitisGen project (www.vitisgen.org). Sixty-five F2 progeny were included for evaluation of berry skin color, for which black (pigmented) and white (nonpigmented) were recoded as 1 and 0, respectively. QTL analysis was performed using the standard interval mapping procedure (EM algorithm) with binary model in the package R/qtl [47] version 1.36.5 for R ver. 3.1.2 [48] and using F2 as the cross type. One thousand permutations at α = 0.05 were executed to calculate the LOD threshold. The additive and dominance QTL effects were estimated taking missing genotype information into account. Missing genotypes were simulated given observed marker data by the hidden Markov model (sim.geno function in R/qtl). Reports were generated for maximum LOD score, 1.5-LOD support interval in cM, and the physical location in the reference genome in Mbp, as well as the percentage of variation explained (R2). The QTL mapping of the quantitative traits SS, MA and YAN was performed in two stages: 1) for each trait independently using composite interval mapping through the package R/qtl [47] version 1.36.5 for R ver. 3.1.2 [48]; and 2) joint QTL mapping for multi-trait trials in a single environment, using the routines QMTQTLSCAN, QMTESTIMATE and QMTBACKSELECT [49] integrated in the statistical package GenStat for Windows 17th [50]. For the three traits, phenotypic information was used from fruit of 63 progeny harvested in 2013. The normality of the distributions was tested by the Shapiro-Wilks tests. Logarithm (Log) transformation was applied for results validation. The QTL detected were similar using either transformed or original data. The markers were coded according to the recommendations of the software used (A, B and H for R/qtl, and as 1|1, 2|2 and 1|2 in GenStat). QTL mapping was executed as described for the berry skin color in R/qtl. The 1.8-LOD support intervals were reported as suggested by “A Guide to QTL Mapping” [51] for mapping traits for an intercross family. In the case of the joint QTL mapping in GenStat, the joint QTL mapping was performed using the phenotypic data for each combination of two traits (MA-SS, MA-YAN and SS-YAN). Cofactors were placed every 1.51 cM along the genome, resulting in placement of 861 cofactors along the genetic map. The model used for the first QTL scan was P dom dom add add þ xidom adom Þ þ TEij , where yij is the yij ¼ m þ Tj þ f 2F ½ðxifadd cadd jf þ xif cjf Þ þ ðxi aj j value of trait j for genotype i, Tj is the trait main effect, F is the set of cofactors, xifadd and


5 / 19


xiadd are the additive genetic predictors of genotype i at the cofactor positions and at the tested position, respectively. The associated effects are denoted by cjfadd and αjadd for cofactors and tested positions, respectively. xifdom and xidom are dominance genetic predictors of genotype i at the cofactor positions and at the tested position, respectively, with associated effects cjfdom, and αjdom. TEij refers to the residual error of the trait j for genotype i. The fit of the unstructured model was iteratively performed through Restricted Maximum Likelihood (REML), using 1000 iterations. The QTL threshold was set at a significance level of 0.05 using the Li and Ji method [52] with minimum cofactor proximity of 50 cM and minimum separation of selected QTLs of 30 cM. Subsequently for the interval mapping and estimation of QTL effects, the model was simpliP dom dom fied to: yij ¼ m þ Tj þ l2L ½ðxiladd aadd jl þ xil ajl Þ þ GTij , where yij is the value of trait j for genotype i, Tj is the trait main effect, xiladd are the additive genetic predictors of genotype i for locus l, and αjladd are the associated effects. xildom are the dominance genetic predictors, and αjldom are the associated effects. The method to fit these models follow the ideas of Malosetti et al. [53] and Boer et al. [54]. Both mixed models consider the genotypes as random terms, while the genetic predictors as fixed effects. The genetic predictors can be considered as genotypic covariables informing the genotypic composition of a genotype at a specific chromosome locus [55]. GTij is assumed to follow a multi-Normal distribution with mean vector 0, and a variance covariance matrix S, which was modeled during the QTL mapping procedure. Finally, to approach the causality between traits MA and SS (causal, reactive, independent, full), a causal model selection test [56] was implemented using the function cmst in the R package qtlhot version 0.9.0 [57].

Results Marker development A total of 291,453 quality single nucleotide polymorphisms (SNPs) were identified in 424 F2 progeny relative to the V. vinifera ‘PN40024’ reference genome, version 12X.v2. Subsequently SNP filtering by minimum read depth and missing data rate yielded 2180 intercross markers with 1:2:1 segregation in the progeny, as determined by a χ2 test of marker segregation. The markers were discovered and coded as hk×hk using a custom perl script (S1 File). Further marker curation and genetic map construction were performed in JoinMap1 4.1, resulting in a final set of 1,449 markers. The workflow for marker development and linkage map construction in the F2 mapping family are presented in Fig 1.

Linkage map construction Linkage analysis using LOD 10.0 grouped the 1449 markers into 19 linkage groups (LG) with a total map size of 2423.9 cM and a map density (average distance between markers) of 1.67 cM (Fig 2). Given the grapevine genome size of about 458.8 Mb [40], the current genetic map is about 0.189 Mb/cM. Based on a comparison of the genetic map to the reference genome (physical map), all linkage groups (except Chr15) had an estimated genome coverage greater than 90%, and the average genome coverage was 95.1% (S2 File). To check marker order, the genetic positions of markers within each LG were compared with physical coordinates on the reference genome (S3 File). Marker order was conserved between the genetic and physical map. Compared to an SSR map previously developed for the same F2 family [29], this GBS map represented a nine-fold improvement in resolution from 14.9 cM to 1.67 cM between markers and an 11% increase in the average genome coverage, from 84% to 95% (Table 1). In addition,


6 / 19


Fig 1. Workflow of marker development and genetic map construction in the V. riparia × ‘Seyval’ F2 family. The left panel shows the five main steps with filtering parameters, and the right panel shows the numbers of markers resulting from each step. doi:10.1371/journal.pone.0149560.g001

there were six gaps greater than 20 cM in the previous SSR map, while the largest gap in the GBS map was 13.5 cM.

Berry color mapping for map validation Berry skin color is controlled by the VvmybA1 gene (Chr02: 14,179,266–14,180,746) [58]. This well-studied trait was used to validate the soundness of the F2 map construction strategy. Berry skin color was recorded for 65 F2 progeny and showed a 51:14 ratio of black:white (pigmented: nonpigmented), which was not significantly different from 3:1 (Chi-squared test, p = 0.80, S4 File), confirming the trait was controlled by one completely-dominant gene. Only one significant QTL, with a peak LOD score of 12.47 (threshold = 3.9), was identified on LG02, and explained up to 90% of the phenotypic variance for berry skin color (Fig 3A). The 1.5-LOD support interval spanned the genetic map at 71.06–90.85cM, corresponding to Chr02 6.3–14.3Mbp and encompassed the VvmybA1 locus (Table 2). The estimated additive effect was 0.5 (a = 0.50), equal to the estimated dominant deviation (d = 0.46), which perfectly matched the one-locus, complete-dominance model and thereby validated the F2 map construction strategy (Fig 3B).

QTL mapping for enological traits QTL mapping of the enological traits YAN, MA and SS was conducted using composite interval mapping (CIM) and the Kosambi mapping function. Each phenotype followed a normal


7 / 19


C hr01 C hr02 C hr03 C hr04 C hr05 C hr06 C hr07 C hr08 C hr09 C hr10 C hr11 C hr12 C hr13 C hr14 C hr15 C hr16 C hr17 C hr18 C hr19 0.00

S1_61235

4.58 6.02 7.22 7.58 8.29 9.50 10.75 11.37 12.96 14.22 14.94 16.75

S1_450298 S1_337751 S1_573072 S1_573176 S1_524396 S1_599235 S1_656290 S1_656299 S1_666584 S1_759430 S1_771037 S1_974740

18.31

S1_1098043

20.73 21.93 22.05 23.73

S1_1505501 S1_1552601 S1_1552591 S1_1785756

27.11 27.23 28.61 29.57 30.91 31.99 32.84

S1_2565156 S1_2565134 S1_2861205 S1_3016043 S1_2958471 S1_3062152 S1_3145439

35.18

S1_3444710

36.82

S1_3587278

39.22

S1_3673664

40.80

S1_3738134

43.72

S1_4096026

48.33

S1_5026255

49.94 51.15 52.86 52.98

S1_5061702 S1_5123842 S1_5605223 S1_5605220

55.60

S1_6116585

57.24

S1_5989383

59.82 61.19 62.04 62.16 63.61 64.81 65.40 66.49 67.88 69.25

S1_6289520 S1_6421612 S1_6450350 S1_6450340 S1_6691294 S1_6883157 S1_6927353 S1_6912355 S1_7244733 S1_7244667

71.79 72.50 72.98 74.29 75.26 75.97 77.05 78.66

S1_8372195 S1_8028456 S1_8381509 S1_8645444 S1_8777783 S1_9218054 S1_9300100 S1_9830551

80.81 82.16 83.15 84.43

S1_10608331 S1_10996335 S1_10830727 S1_11272101

86.00

S1_11168598

87.56

S1_11378530

92.16

S1_12515801

100.19

S1_17670419

102.66 102.78

S1_18502185 S1_18502208

105.82

S1_19145591

109.56

S1_19658214

112.54

S1_20356240

115.93

S1_21031367

125.43

S1_21970357

0.00

S2_99114

2.91

S2_183880

7.38 7.62 7.87 8.84 9.08 10.40

S2_474669 S2_474699 S2_474702 S2_475167 S2_475182 S2_539315

13.65

S2_715118

21.61

S2_1295510

23.57

S2_1406827

29.55

33.95 36.03 37.56 38.82 40.45 41.96 42.63 43.39 44.35 45.67 47.47 48.58 49.86 51.00 51.59 51.88 52.02 53.15 53.86 54.95 55.85

S2_1905793

S2_2196035 S2_2435582 S2_2563264 S2_2592172 S2_2794425 S2_2957597 S2_2906134 S2_2974676 S2_2994294 S2_3293020 S2_3617261 S2_3629132 S2_3674984 S2_3794440 S2_3953244 S2_3835305 S2_3835280 S2_3983797 S2_3983708 S2_4102314 S2_4240162

58.49

S2_4428879

60.85 62.13

S2_4567948 S2_4650238

66.07

71.06

S2_5352505

S2_6360259

0.00 1.01 2.98 3.36 4.23 6.09 7.25 8.09 9.66

S3_41418 S3_43405 S3_532967 S3_532978 S3_624278 S3_1001036 S3_1023006 S3_990949 S3_1077230

11.27

S3_1092472

13.77 14.61 15.81 17.16 17.69 19.27 20.64 21.99

S3_1484998 S3_1493744 S3_1493546 S3_1564529 S3_1564532 S3_1627732 S3_1629495 S3_1675012

24.91 26.89 28.29 28.96 30.04 31.32 32.59 33.55 34.42 35.01 35.83 36.92 37.62 38.84 40.17 41.01 41.85 42.92 43.52 43.96 45.31 45.91 46.74 47.62 48.29 49.54 51.55 52.99 54.21

S3_2117004 S3_2247132 S3_2110082 S3_2108601 S3_2344322 S3_2466095 S3_2346547 S3_2541773 S3_2421441 S3_2421437 S3_2404144 S3_2413800 S3_2452475 S3_2730328 S3_2823451 S3_2878868 S3_3098043 S3_3151577 S3_3204618 S3_3211065 S3_3631916 S3_3921459 S3_3675429 S3_4222089 S3_4239537 S3_4520612 S3_4877432 S3_4935304 S3_4950290

23.64

S4_1475521

25.51 27.60 28.91 29.56 30.16 30.27 31.65

S4_1475655 S4_1800675 S4_1631883 S4_1522289 S4_1532811 S4_1532789 S4_1856634

34.45 35.47 36.55 37.69

S4_2211842 S4_2283799 S4_2112469 S4_2253063

39.42

S4_2408273

55.74

S4_4331738

57.77 58.97

S3_5417721 S3_5605883

58.00 59.44

63.43

S3_5884897

66.80

S3_6200440

72.57

S2_6196895

72.08

S3_7105462

74.52 75.52

S2_6967060 S2_6897782

74.00 75.46

S3_7193302 S3_7382147

80.62 82.05

S3_8665103 S3_8895634

78.27

S2_7651502

80.13

S2_8076520

82.37 83.01 84.19 85.40

S2_8746140 S2_8713657 S2_8874350 S2_9126568

87.51 87.89

S2_10795173 S2_10795172

90.85

S2_14300736

92.41

S2_14108196

93.99

S2_17017148

95.89 96.01 97.48

S2_17839875 S2_17839879 S2_17854965

100.31 101.16 101.93 103.04

S2_18507119 S2_18456742 S2_18455557 S2_18702143

90.07 90.80

S3_14376521 S3_14322374

94.50

S3_16774440

98.78

S3_18697436

102.21

S3_19185179

0.00

S4_496922

6.29

S4_417190

9.31 10.73 11.82 12.67 13.66

S4_833538 S4_936532 S4_1005643 S4_1048504 S4_1063770

15.54

S4_1136396

17.42 18.81 19.16 20.12 21.56

S4_1262613 S4_1274086 S4_1274066 S4_1281341 S4_1336335

42.36

S4_2873325

43.97 45.05 46.04

S4_2933682 S4_3131726 S4_3228910

47.90

S4_3284361

49.43 50.51 51.84 52.80 54.13

S4_3602406 S4_3632240 S4_3818638 S4_3781992 S4_4192726

0.00 1.01

S5_58268 S5_221245

2.79 3.68

S5_430550 S5_573962

6.84 8.28 9.85 10.56 11.16 12.13

S5_844700 S5_1057931 S5_1221266 S5_1161567 S5_1162334 S5_1162354

16.19

S5_2591080

18.16

S5_2201989

20.21 21.24

S5_2388377 S5_2500900

23.57 24.59 25.40

S5_4419861

40.70

S5_4898738

42.70 42.83

S5_4540221 S5_4540238

41.99 42.61

S6_5076310 S6_5065181

44.81 46.18

S5_4617260 S5_4631768

46.18

S6_5498696

48.47 48.59 49.67

S5_5111572 S5_5111562 S5_5442628

48.73

S6_5589380

S4_5755073

73.37 74.68 75.54 76.39 77.97 78.69 79.65 81.24 82.15 83.59

S4_5921474 S4_6021282 S4_6033723 S4_6033639 S4_6389633 S4_6609588 S4_6574395 S4_6702849 S4_6751985 S4_7027174

85.46

S4_7599224

87.40

S4_7806435

108.78 109.86 110.46 112.18 112.80

S4_15628172 S4_15156984 S4_15358133 S4_16592523 S4_16771681

117.63 119.01

S4_17950348 S4_17966025

122.19

S4_18371526

124.70

S4_18748613

127.10 127.94 128.79

S4_18940370 S4_18945961 S4_18963924

130.69

S4_19300635

132.31 133.51 134.35

S4_19502024 S4_19401514 S4_19558530

136.43 137.84

S4_19736889 S4_20016288

139.46

S4_20146063

142.60

S4_20396066

144.12 145.33 146.29

S4_20376486 S4_20476409 S4_20482368

148.52 149.98

S4_20619675 S4_20708456

151.51 152.90 154.11 154.83

S4_20834667 S4_21125912 S4_21070589 S4_21192525

157.10

S4_21609026

158.90

S4_21780964

162.34 163.21 164.78 165.57 166.66 168.02

S4_22841476 S4_22856230 S4_23253534 S4_23556631 S4_23839694 S4_23891111

169.59

S4_24028647

171.18 172.74 172.99 174.32 175.56

S4_24242954 S4_24448098 S4_24447249 S4_24561501 S4_24606925

S6_3524817

65.00 65.71 67.26

S5_7569584 S5_7951955 S5_8195964

68.83

S5_8471721

70.46 71.48 72.56 73.52

S5_8866891 S5_8842307 S5_8933721 S5_8933647

75.96 76.91

S5_9694212 S5_9694146

79.34 80.25

S5_10302981 S5_10710229

83.70 84.73 86.22 86.35 88.22 89.36 89.60 90.92 92.04

S5_11920108 S5_12614447 S5_12615053 S5_12615045 S5_13958752 S5_14718188 S5_14718162 S5_15323715 S5_15495055

95.34 95.58 96.20 97.49 98.99 99.39 101.63 101.75 103.07 103.92

S5_17868531 S5_17868530 S5_17652900 S5_17910491 S5_17910565 S5_17910544 S5_18521423 S5_18521398 S5_18497582 S5_18379491

107.96 108.81 110.01

S5_20508317 S5_20829356 S5_20877514

113.54 114.38 115.86 116.83

S5_21855287 S5_21855304 S5_21771486 S5_21730188

118.61 120.00

S5_22788793 S5_22762976

121.75 122.83 124.22 125.15 126.42

S5_23164926 S5_23066867 S5_22556804 S5_22425209 S5_22631511

128.47 128.59

S5_23224391 S5_23224401

131.29

S5_23612814

133.95 136.25 136.37 136.96 137.68 139.00 141.14 142.68 143.71 144.67 145.63 146.46 146.58 148.87 148.99 149.83

S5_23871533 S5_24031772 S5_24031761 S5_24080724 S5_24031809 S5_24280625 S5_24319861 S5_24729888 S5_24843159 S5_24816875 S5_24871426 S5_25073824 S5_25073804 S5_25438857 S5_25438869 S5_25494880

54.71 55.92 56.76 57.59

S6_6497323 S6_6491927 S6_6499237 S6_6493525

61.08

S6_7234231

62.95 64.04

S6_7343698 S6_7421599

65.70 67.53 68.96 70.24 71.32 72.22 73.54 75.23 76.25 77.26 78.37 79.21 80.05 80.78 80.90 82.04 82.45 84.08 86.03 86.98 87.74 88.50 89.05 90.36 91.56 92.03 92.87 93.70 94.06 94.53 95.25 96.57 97.16 98.40 98.77 100.22 101.22 103.13 105.12 106.34 107.68 108.89 110.40 112.29 114.11 115.49 116.65 118.02 119.03 120.07 121.41

S6_7661178 S6_7723596 S6_7723710 S6_7985435 S6_8129979 S6_8056477 S6_8276066 S6_8358059 S6_8833805 S6_9327689 S6_11692258 S6_11606795 S6_11920562 S6_11876418 S6_11963135 S6_13471231 S6_13625092 S6_14045591 S6_14522242 S6_14263150 S6_14262956 S6_15006120 S6_15122181 S6_14977355 S6_15476231 S6_14871350 S6_15857118 S6_15778007 S6_15783761 S6_15783752 S6_16115245 S6_16634256 S6_17440350 S6_17635096 S6_17635115 S6_17445983 S6_17515555 S6_18144892 S6_18730922 S6_18946821 S6_19117457 S6_19048262 S6_19521142 S6_19923276 S6_20201501 S6_20302700 S6_20392134 S6_20425090 S6_20601652 S6_20918375 S6_21030267

123.86 125.11 125.70 126.23 127.47

S6_21593950 S6_22059378 S6_22246199 S6_22228284 S6_21919991

129.66

S6_21625792

0.00 0.98 1.69 3.02 4.34 5.78 7.28 8.45 9.39 9.99 11.20 12.65 13.73 14.57 15.78 16.39 17.24 18.56 19.52 20.59

S8_324958 S8_293230 S8_117400 S8_721128 S8_1353107 S8_2791177 S8_3002378 S8_3793471 S8_4284123 S8_4284080 S8_4203272 S8_3793545 S8_5252235 S8_4963325 S8_5458379 S8_5877926 S8_7219197 S8_4596364 S8_5958731 S8_7486240

S7_1751381

23.88

S8_8735489

25.64

S7_1983175

25.96

S8_9778757

28.01

S7_2309722

29.96

S7_2233180

29.50

S8_10343142

32.54 33.35

S7_2923127 S7_2925238

32.86

S8_10660397

35.26

S7_3299521

34.56

S8_10800727

36.93 38.12

S7_3836234 S7_3843535

39.79 40.84

S7_3965728 S7_4079352

43.53

S7_4373880

23.73 27.07

37.59

71.41

S4_11235216 S4_11594980 S4_12023566 S4_12378387 S4_12378356 S4_13165925 S4_14441447 S4_14520796

S6_2854997

S6_4258443

S4_5402978

S4_10225996

S6_2546332 S6_2755982 S6_2599021 S6_2599048

21.29

35.85

68.27

S4_10226058

15.27 17.06 18.34 18.72

S5_4235745

S5_6920131

S4_9425560

S6_1894535

35.47

S5_6531777 S5_6531868 S5_6581873 S5_6617966 S5_6617972

S4_8483946

10.75

S6_3857140

61.08

S4_9013707

S7_303849 S7_516368 S7_785302 S7_849517 S7_860258 S7_1227353 S7_1235082 S7_1225672 S7_1231413 S7_1405946 S7_1408772 S7_1408783 S7_1430460 S7_1423130 S7_1435517

32.09

54.88 56.08 57.43 58.63 58.75

98.40 99.48 101.03 102.30 103.22 104.13 105.49 106.57

S7_233994 S7_236098

4.75 6.08 7.47 8.37 9.21 10.89 11.61 12.37 13.66 15.94 17.16 17.28 18.36 19.08 20.21

S5_3370451

S4_4796042

94.79

1.75 3.04

S5_3642668 S5_3642669

S4_4971246 S4_4973663 S4_5019270 S4_5203161

96.35

S7_91163

S6_908236

28.68

S4_4708744 S4_4639589

92.24

0.00

S6_745440

4.09

31.30 31.43

61.13

89.10

S6_226453

2.12

S5_2734909 S5_2724319 S5_2780288

62.78 63.77 64.73 65.93

90.70

0.00

45.09

S7_4417741

46.76

S7_4348828

48.67

S7_4538697

50.76

S7_4643400

52.67 53.87

S7_4710219 S7_4773542

55.82 56.82

S7_4879939 S7_5010810

58.46

S7_5156168

60.87 60.99 62.28 64.13

S7_5245421 S7_5245411 S7_5266672 S7_5380011

66.52 67.72 69.04 70.00 71.01

S7_6338231 S7_6634749 S7_6634646 S7_6634661 S7_6652023

73.26 74.37

S7_7022024 S7_7022084

76.87 77.76 78.79 81.46 81.93 82.77 84.03 84.28 85.60 86.69 86.93 88.12 90.60 91.45

S7_7379085 S7_7529804 S7_7755420 S7_9420879 S7_9420884 S7_9418460 S7_9416126 S7_9416158 S7_10471196 S7_10587288 S7_10587303 S7_10079728 S7_15546340 S7_14622489

93.42 94.15 94.87 95.46 96.31 97.40 98.61 100.44 101.85 103.00

S7_16422022 S7_17314422 S7_16987038 S7_17234461 S7_17307914 S7_17610238 S7_18080009 S7_18819915 S7_19090324 S7_18701379

38.71 40.15

S8_11176988 S8_11360063

41.99

S8_11444114

44.13 45.04

S8_11536729 S8_11523467

46.74 47.90 49.28 50.15

S8_11631412 S8_11642930 S8_11785640 S8_11791483

53.78

S8_12271849

59.66

S8_13385739

61.68 61.94

S8_13359043 S8_13359036

67.07 68.11

S8_13783126 S8_13784711

71.10

S8_14136237

73.72

S8_14353985

77.80

S8_14747705

81.23

S8_15111633

S8_15729357

87.32

S8_16041820

92.40 92.53 93.98 95.56 96.58 98.29 99.50 100.40 101.06 102.07

S8_17028351 S8_17028353 S8_17257633 S8_17208839 S8_17054594 S8_17440416 S8_17766225 S8_17766135 S8_17589064 S8_17545275 S8_18216438 S8_18216754 S8_18216903 S8_18216641 S8_18216599 S8_18168099 S8_18263859 S8_18329635 S8_18796228 S8_19239337 S8_19539539 S8_19524960 S8_19524949 S8_19483434 S8_19550404 S8_19001342 S8_19001348 S8_19192550 S8_19955887 S8_20067611 S8_20137705 S8_20445101 S8_21105472 S8_20909403 S8_21059734 S8_21372950 S8_21406499 S8_21708888

104.83

S7_19122795

106.85 108.16

S7_20565550 S7_20473403

114.95

S7_20907108

123.36

S7_22038492

125.77

S7_22156653

127.75 128.22

S7_22486245 S7_22486215

130.46 131.85

S7_22613627 S7_22649363

133.78

S7_22735436

136.42

S7_22940661

136.41

138.57 139.71

S7_23132810 S7_23125517

138.02

S8_21823877

139.64

S8_21929790

141.44

S7_23208963

143.62 144.85

S7_23670151 S7_23667371

141.50 142.75

S8_22140676 S8_22239521

144.83

S8_22458965

148.70

S7_23921831 S7_23958963

152.22

S7_24029698

159.15

S7_24822970

166.01

S7_25866846

173.40

S7_27019046

S9_248117 S9_295085 S9_295109 S9_382705 S9_621587 S9_1139405 S9_1163621 S9_1202538 S9_1337702 S9_1468311 S9_1519553 S9_1518021 S9_1518031 S9_1911192 S9_1900874 S9_1961989 S9_1961993 S9_2191771 S9_2309792 S9_2360223 S9_2451554 S9_2457295 S9_2906757 S9_2901231 S9_3040244 S9_2920669 S9_3079786 S9_3079788 S9_3107458 S9_3378182 S9_3468893 S9_3468875 S9_3741994 S9_3741973 S9_3870484 S9_4086424 S9_4352513 S9_4352496 S9_4435047 S9_5446175 S9_5446172 S9_5375354 S9_5375340 S9_5740727 S9_5615543 S9_5756639 S9_6230049 S9_6190338

57.08

S9_6990548

61.41

S9_7547166

72.13

S9_9648613

74.28 75.50 76.84

S9_10501451 S9_10510613 S9_10733549

83.33 85.07

103.98 105.45 106.66 107.74 108.34 109.37 110.41 111.52 113.12 114.54 115.22 116.42 116.54 117.57 118.33 119.66 119.79 120.92 122.40 124.35 125.58 126.83 128.23 129.11 130.31 131.69 132.97

150.30

0.00 1.55 1.79 2.99 5.51 8.36 10.06 11.14 13.79 14.53 15.97 17.37 17.49 19.69 20.29 21.26 21.51 23.43 24.98 25.94 26.58 27.49 28.93 29.77 30.36 31.00 32.04 32.16 33.17 34.54 35.62 35.74 36.81 36.93 38.33 39.32 40.77 41.04 42.57 44.87 44.98 45.58 45.82 46.78 47.68 48.63 50.40 51.37

S9_13661662

90.14 90.87 92.10 92.22 93.21 94.51 96.09 97.55 98.39

S9_18475530 S9_18392826 S9_18487675 S9_18487656 S9_18597650 S9_18875795 S9_19000027 S9_18999969 S9_18884162

100.34 101.72

S9_19481176 S9_20088952

104.81 108.75 109.10 110.31 112.25 113.15 114.02

S9_21136414 S9_21899594 S9_21899595 S9_21899644 S9_22885291 S9_22795298 S9_22639647

0.00 1.47 2.86 4.48 5.22 6.25 7.18 8.51 8.86 10.25 11.95 13.16

S10_34153 S10_592094 S10_425096 S10_425046 S10_429263 S10_498103 S10_371175 S10_402217 S10_402218 S10_592175 S10_1036341 S10_1013405

14.75

S10_1362195

16.23 17.44 18.75

S10_1383170 S10_1377150 S10_1472990

32.26

S10_3606712

36.44

S10_4177739

40.70

S10_4583458

0.00

S11_279002

0.00

S12_236794

2.55

S11_543331

5.03

S11_637120

3.39 4.50 4.74

S12_556288 S12_636820 S12_636844

6.61

S11_637003 8.07

S12_1419004

10.04 11.05

S12_1520859 S12_1609761

13.39

S12_1685433

15.08 16.87 17.24 18.69 19.07

S12_1768883 S12_1850655 S12_1850650 S12_1957046 S12_1957076

22.81 24.76 25.39 25.74 26.82 28.85 29.13 29.86 31.85 32.46

S12_2645328 S12_3026118 S12_3110027 S12_3110039 S12_3179114 S12_3328879 S12_3328891 S12_3310532 S12_3754143 S12_3754160

9.82

14.05

S11_1126416

16.16 16.69

S11_1163923 S11_1163836

22.41

S11_1973162

24.34

S11_2056883

27.95

S11_2440831

30.63

S11_2543400

32.50

S11_2601023

34.64

S11_2635820

34.59

S12_4004658

37.44

S11_2824566

37.32

S12_4625727

39.57

S12_4759679

45.08 45.95 47.26 50.04 51.30 52.97 53.23

S10_5599103 S10_5599901 S10_6012906 S10_6012894

55.30

S10_6554504

56.79

S10_6554496

62.26

S10_7271268

65.07 66.09 67.22 68.29 68.53 69.60 69.84 70.34 71.63 72.82 73.06 76.08 76.35

S10_8100733 S10_8091011 S10_8172593 S10_8101899 S10_8101892 S10_8400773 S10_8400795 S10_8505086 S10_8694580 S10_8789892 S10_8789856 S10_9799045 S10_9799090

79.69

S10_10408546

82.95 83.54

S10_10801568 S10_10733313

85.87

S10_11553529

94.14

S10_16926564

97.53 99.41 99.66 100.03 101.61

S10_21780463 S10_22647189 S10_22647217 S10_22601505 S10_22315814

S11_907711

S11_3485916 S11_3461670 S11_3557289

51.31 52.18

S11_4016284 S11_4104192

58.45

S11_4722239

64.14 65.59 66.92

S11_5249786 S11_5265819 S11_5285218

69.99

S11_5898318

74.42 74.77 76.09 77.42 78.62 78.85 79.33 80.30 81.52 82.49 83.46 84.66 85.92 87.33 87.57 88.60 89.12 89.72 89.83 90.91 91.99 93.30 95.00 95.72 96.08 96.91 98.89 101.44 103.61 105.53 106.53

S11_6607285 S11_6607300 S11_7006742 S11_7260047 S11_7658208 S11_7658205 S11_8155332 S11_7860814 S11_7696089 S11_8649508 S11_8997540 S11_8490934 S11_8648407 S11_9676211 S11_9676227 S11_9838587 S11_10309754 S11_10200818 S11_10200810 S11_11351426 S11_12378540 S11_13021165 S11_12578258 S11_12610633 S11_13189428 S11_13317144 S11_16994391 S11_18591959 S11_19405691 S11_19921280 S11_19776637

42.27 43.47

S12_5131553 S12_5119441

46.82

S12_5613074

48.33

S12_5529125

53.24 54.45 55.45

S12_6382037 S12_6546333 S12_6491869

57.96

S12_6767109

60.12

S12_6855572

62.28

S12_6892734

65.08 66.38

S12_7358079 S12_7377974

68.16 69.38 70.64

S12_7475656 S12_7527084 S12_7502017

73.14

S12_8113436

0.00 1.38 2.58 3.18 5.25 6.54 7.75 8.11 9.44 10.03 11.10 11.94 12.46 12.85 13.76 15.65

S13_242953 S13_162274 S13_305342 S13_356925 S13_540663 S13_565207 S13_606086 S13_606112 S13_671193 S13_671370 S13_927335 S13_937599 S13_936756 S13_936730 S13_965491 S13_1305382

18.14 18.93 20.42 21.64 21.88

S13_1678743 S13_1685606 S13_1794111 S13_1806932 S13_1806912

24.27 25.37 26.78 27.81 27.93 28.89 30.66 32.09

S13_2205259 S13_2230198 S13_2272918 S13_2272929 S13_2272945 S13_2343651 S13_2438004 S13_2688589

34.63 35.72 37.18

S13_3077993 S13_2953718 S13_3270856

39.51

S13_3450532

41.27 42.72 43.84 44.51 45.23 46.63

S13_3646068 S13_3795594 S13_3906951 S13_4005097 S13_3966084 S13_4338318

52.90 53.68 54.58 54.70 55.05 55.65 55.77 56.41 58.65 61.23 63.03 64.54 65.31 66.77 67.60

S13_5276826 S13_5371521 S13_5255723 S13_5255677 S13_5370150 S13_5346318 S13_5346316 S13_5378859 S13_5623472 S13_5764308 S13_6125095 S13_6727708 S13_6736188 S13_7174866 S13_7405457

70.47

S13_7795575

75.15

S12_8202900

74.04

S13_9506557

77.58 78.20 79.22

S12_8656740 S12_8738784 S12_8818252

77.61

S13_11400257

81.17

S12_8935262

81.63 82.78 82.90 83.97 84.09 85.53

S13_14380064 S13_14509664 S13_14509641 S13_14644898 S13_14644906 S13_15082975

88.23

S13_16150992

89.83

S13_16718443

91.66 93.56 94.99 96.11 96.83 97.08 99.01 99.52 99.89

S13_16733953 S13_18212211 S13_18394324 S13_18972960 S13_18815025 S13_18815052 S13_21663896 S13_21357147 S13_21357132

102.98 103.82 104.66 106.24

S13_20488089 S13_20346548 S13_22023111 S13_22492642

109.78 111.15 112.03

S13_24768993 S13_25221220 S13_25156621

82.94

S12_8983011

84.98

S12_9587137

87.04

S12_9730944

88.97 90.41 91.87 93.55 94.75 95.46 96.15

S12_12148332 S12_12220260 S12_12726664 S12_12773875 S12_12833686 S12_12974761 S12_12833642

100.29

S12_15271043

102.94 103.66 104.55

S12_16561904 S12_16438394 S12_16418334

109.77

S12_19463132

113.73

S12_21068579

116.13 117.44

S12_21431241 S12_21429987

120.49 121.77 122.41 123.45

S12_22565016 S12_22409192 S12_22409191 S12_22583732

126.12 127.50 128.66 129.40

S12_23473436 S12_23853447 S12_23964982 S12_24257058

116.85

S13_26234024

119.70

S13_26605683

121.27

S13_26774283

122.79 123.87

S13_27044671 S13_26988211

127.90 129.11

S13_27775343 S13_27780950

130.94

S13_28078667

132.50

S13_28149587

140.36 141.57

S13_29053383 S13_29035062

143.05

S13_29053915

0.00 1.10

S14_307761 S14_335336

5.51

S14_996837

9.31

0.00

S15_202858

2.05 3.53 4.41 5.38 6.59 7.62

S15_1187342 S15_1442037 S15_1644882 S15_1566689 S15_1644501 S15_1931143

11.24 12.60

S15_5844417 S15_6987339

S14_1578388

14.81

S14_3153680

18.39 18.51 19.54 21.33

S14_3537895 S14_3537870 S14_3596972 S14_4330259

24.31 26.39 26.86 27.68 29.15 29.63 30.47 32.54 32.66 34.58 36.14 36.26

S14_5049385 S14_5769600 S14_5737577 S14_5804821 S14_6071298 S14_6003418 S14_6071668 S14_6783754 S14_6783773 S14_7249434 S14_7565247 S14_7565248

40.04

S14_9109022

42.19

S14_10187131

44.51

S14_11577226

46.38 48.04 48.56

S14_12546097 S14_11011400 S14_11011429

52.53 52.76 53.60

S14_15718793 S14_15718807 S14_15718811

55.91

S14_16489495

62.20

S14_18307597

66.23 68.32 68.68 68.92

S14_19528236 S14_20102915 S14_20102957 S14_20102935

74.33 75.96 76.59 77.90 78.02 79.39

S14_21775987 S14_21836090 S14_21868784 S14_21640998 S14_21640995 S14_21836159

82.90

S14_22044025

86.81 87.93

S14_22696881 S14_22847899

92.22

S14_23352956

95.26

S14_23909530

98.00

S14_24204392

101.15 102.00 102.89 103.37 103.61 104.81 106.51 108.28 109.33 110.70 112.29 113.56 114.44 115.84 117.05 118.01 119.11 119.83 120.30 120.84 121.76 122.12 122.65 123.56 124.96 127.02 127.29 127.93 129.07 130.08 130.62 132.90 134.26 136.00 137.91 138.72 139.85 141.53 142.68 143.80 144.81 144.95 146.17 147.14 148.45

S14_24792599 S14_25069455 S14_24674362 S14_24980155 S14_25002430 S14_25060615 S14_25092532 S14_25236227 S14_25210247 S14_25355087 S14_25628210 S14_25535909 S14_25689687 S14_25984163 S14_26263802 S14_26224265 S14_26302319 S14_26235789 S14_26502180 S14_26532459 S14_27118841 S14_27118879 S14_27118893 S14_27170394 S14_27100557 S14_27650708 S14_27650732 S14_27418355 S14_27925228 S14_27846258 S14_28215268 S14_28527609 S14_28669910 S14_28949178 S14_29313971 S14_29213966 S14_29222719 S14_29564092 S14_29685950 S14_29658531 S14_29686448 S14_29686438 S14_29896121 S14_30108296 S14_30072192

15.16

S15_9666397

16.74 18.12

S15_8117154 S15_10186072

19.83 21.30 22.57 24.08 25.08 26.28

S15_10293273 S15_10854347 S15_10640514 S15_11136588 S15_11289343 S15_11310734

28.72

S15_11563406

31.06

S15_12566419

32.64

S15_12655908

0.00 1.49 2.42 3.45 5.07 5.91

S16_837377 S16_360918 S16_180419 S16_157508 S16_507134 S16_455568

8.85

S16_1190932

12.05

S16_1746314

14.01

S16_1785606

17.06

S16_2385735

19.16 20.24

S16_2628676 S16_2628785

21.80

S16_2668936

23.51

S16_2789979

27.54

S16_2996955

30.91 32.51 32.63 34.08

S16_6006888 S16_6228053 S16_6228083 S16_6692998

37.46

0.00

S17_217269

3.24 3.94

S17_541171 S17_580505

6.82

S17_713864

12.55 13.75 14.23

S17_1192720 S17_1205741 S17_1205747

17.12

S17_1377163

23.30

S17_2144209

26.67

S17_2470952

31.91

S17_3204678

34.45

S17_3247438

S16_9192332

44.22 44.34 45.45 46.95 48.12 48.69 49.53 51.93 52.43 53.02 54.99 56.20 57.28 58.88 60.61 61.70 62.42 63.13 64.99 65.10 65.59 66.26 67.55 69.15

S16_13661616 S16_13661625 S16_13813688 S16_13656858 S16_13925289 S16_14019789 S16_14618354 S16_15737890 S16_15931047 S16_15931236 S16_16205220 S16_16351096 S16_16350781 S16_16501043 S16_17131628 S16_17223284 S16_17325475 S16_17312507 S16_17577861 S16_17577879 S16_17577830 S16_17577753 S16_17699787 S16_18248907

71.16

S16_18488520

73.52 74.99 75.14 76.11 76.93 77.97

S16_18789936 S16_19531279 S16_19531280 S16_19448504 S16_19431785 S16_19590930

80.62 81.88 82.85 84.12 85.40

S16_20655873 S16_20848074 S16_20759868 S16_20813677 S16_20878346

88.27

S16_22243805

92.85 93.66 94.13 95.70

S16_22599149 S16_22586446 S16_22625111 S16_22742044

99.00 99.24 100.07

S16_22920672 S16_22920660 S16_22920762

102.93

S16_23083281

105.61

S16_23229813

108.79

S16_23286165

113.33

S16_23550586

43.31 44.39

S17_4200704 S17_4143051

48.68

S17_4483655

51.13

S17_4700429

54.67 55.83 56.99 58.47

S17_4956753 S17_4956656 S17_4983574

64.43 66.20 69.03 71.11 73.99 74.36 76.76 78.19 79.10 80.42 81.49 81.97 83.29 84.60 85.60 85.86 87.50 88.54 90.14 90.38 92.19 93.44 93.69 94.46 96.12 96.25 97.74 99.68 101.14 102.90 103.60 104.17 104.72 105.65 106.97 108.68 109.06 110.45 111.55 112.59 113.84 114.92 115.46 115.73 116.01 116.70 117.65 118.61 119.66 120.64 120.76 121.03 121.59 122.43 122.79 123.50 124.46 125.54 126.14 127.51 128.18

S17_5885473 S17_6004065 S17_6278519 S17_6321906 S17_6744824 S17_6744814 S17_7022563 S17_7182575 S17_7103581 S17_7125499 S17_7182653 S17_7182671 S17_7294206 S17_7383782 S17_7646222 S17_7646227 S17_7769058 S17_7749617 S17_8691136 S17_8691103 S17_9014107 S17_9197125 S17_9197119 S17_9208602 S17_9959959 S17_9959964 S17_9855033 S17_10273498 S17_10144707 S17_9715176 S17_9208836 S17_9687176 S17_9687165 S17_9654615 S17_10441825 S17_10615068 S17_10615063 S17_10642186 S17_10780033 S17_10872488 S17_10839875 S17_13760106 S17_12418863 S17_12418896 S17_12418905 S17_14979833 S17_11143959 S17_11740499 S17_11334231 S17_14858272 S17_14858301 S17_14309157 S17_15044163 S17_16556959 S17_16878362 S17_17361170 S17_16959424 S17_16959946 S17_17645065 S17_17294284 S17_17294290

S17_5248835

0.00 0.87 1.83 3.25 4.36 5.43

S18_262650 S18_348524 S18_332033 S18_496563 S18_579775 S18_616615

0.00

S19_279301

2.25 3.25

S19_52839 S19_19204

5.07 6.03 7.33

S19_305497 S19_373415 S19_350268

S18_1494244 S18_1535620

12.58 13.69 14.77

S19_1680257 S19_1720631 S19_1741182

17.64 19.17 19.56 21.37 21.96

S18_2438568 S18_2451147 S18_2451141 S18_2677393 S18_2677405

17.44 19.16 20.00 21.15 21.67

S19_1896354 S19_2084205 S19_2109611 S19_2485442 S19_2368640

24.90

S18_3163189

26.76

S18_3102047

24.31 24.58 26.24

S19_3049736 S19_3049775 S19_3168101

29.51 30.07 31.30 32.51

S18_3387646 S18_3434842 S18_3649790 S18_3576043

27.85 28.83

S19_3286313 S19_3413391

31.68

S19_3856392

7.21

S18_907657

9.08 10.40

S18_1133264 S18_1279546

12.54 13.60

34.17

S18_3929237

35.72 36.86 38.39 38.90 40.63 41.71

S18_3968397 S18_4017339 S18_4183027 S18_4182999 S18_4382117 S18_4493267

35.21

S19_4232836

36.84

S19_4808967

39.31

S19_5404369

43.84 45.22 46.08

S18_4728336 S18_4701144 S18_4701680

48.56 49.82 50.87

S18_5173462 S18_5351908 S18_5428581

52.51 53.35 54.53 56.29 57.25 58.10 58.71

S18_5579663 S18_5497681 S18_5488867 S18_5914778 S18_6217819 S18_6031168 S18_6031152

43.42 44.70 45.84 47.20 48.13 49.72 50.80 51.76 52.48 53.68 55.24 55.48 57.04 58.49

S19_5927881 S19_5995532 S19_5928151 S19_5928214 S19_5920970 S19_6186424 S19_6179027 S19_6454507 S19_6414363 S19_6778213 S19_6699171 S19_6699192 S19_6813959 S19_6943544

60.89

S18_6479860

60.00

S19_6935550

62.76

S18_6511647

64.65

S18_6721408

68.03

S18_7075016

61.77 62.61 63.85 65.13 66.59 67.79 68.03

S19_7054837 S19_7255783 S19_7340217 S19_7455167 S19_7836327 S19_8597548 S19_8597551

70.89 72.32 72.44 73.54 74.62

S18_7329580 S18_7496114 S18_7496112 S18_7610682 S18_7622618

77.67

S18_8335796

77.10 77.36

S19_13230977 S19_13230959

79.70

S19_17409591

91.93

S19_20623409

80.31

S18_8577418

82.43 84.00 85.07 86.06 87.32

S18_8736936

88.97

S18_9692770 S18_10015467

90.44 91.92 93.06 94.33 95.23

S18_8985216 S18_9010231 S18_9252944 S18_9250073

S18_10210581 S18_10396343 S18_10296559 S18_10387891

97.36 98.80

S18_10786925 S18_10821503

100.63

S18_10863926

102.21

S18_10887948

104.25

S18_11206896

106.18 107.24

S18_11381025 S18_11386981

109.69 110.91

S18_11494321 S18_11497715

112.74

S18_11730982

114.32 115.54 116.64 118.02

S18_11969266 S18_12276238 S18_12298983 S18_12300318

119.56

S18_12137323

121.61

S18_12609810

123.37

S18_12707445

126.21

S18_13135217

127.89 129.29 130.40

S18_13462577 S18_13447596 S18_13411775

132.13 134.24 135.29 135.99 136.40 137.20 138.53

S18_13366031 S18_14160612 S18_14524094 S18_14486022 S18_14486026 S18_14382672 S18_14407308

141.54 141.91

S18_14225672 S18_14225706

144.53

S18_14787621

146.14 147.60 148.25 149.22

S18_17399392 S18_19961860 S18_21535731 S18_20428327

151.82 152.66 153.26 154.26 156.03

S18_22284180 S18_22652878 S18_22663315 S18_22843741 S18_23674610

158.08 159.71 161.04 161.52 162.17

S18_25043028 S18_25489438 S18_26392343 S18_26392346 S18_26379100

165.69 165.95

S18_27784874 S18_27784872

168.27

S18_28125871

178.81

S18_31160355

186.20 186.93

S18_33388874 S18_33406870

195.45

S18_34314381

96.17

S19_22763529

98.54 100.43 100.55 101.45

S19_23054883 S19_23232669 S19_23232668 S19_23252636

105.91 107.28

S19_24491792 S19_24553834

Fig 2. Linkage map of F2 mapping family derived from V. riparia × ‘Seyval’. The linkage map consists of 1,449 markers with a total map size of 2423.9 cM. The 19 chromosomes are labeled at the top of each linkage group. Genetic positions (cM) and physical locations (‘PN40024’ reference version 12X.v2) are listed to the left and right of each chromosome, respectively. doi:10.1371/journal.pone.0149560.g002

distribution (S4 File). A QTL corresponding to YAN was identified on LG07 with a peak at 100.44 cM (LOD = 6.48; Fig 4A). This YAN QTL spanned a physical interval from 17.0–20.6 Mbp (1.8-LOD support interval) and explained 22.8% of phenotypic variance (Table 2). The additive effect of the QTL was negative, indicating that higher concentrations of YAN came from the grandparent 1, V. riparia. For MA, a single QTL was observed on LG06 with a peak at 70.24 cM (LOD = 6.24). This MA QTL contained a relatively narrow 1.8-LOD support interval from 7.3–8.4 Mbp and explained 26.2% of the phenotypic variance (Fig 4B, Table 2). In contrast to YAN, the positive additive effect of MA refers to higher concentration in the grandparent 2, ‘Seyval’. For the ratio of MA/SS, a single QTL was identified on LG06, peaking at 32.10 cM (LOD = 5.93). This MA/SS QTL spanned a relatively narrow interval between 3.5 and 4.3 Table 1. Comparison between the previously published SSR-based map and the GBS-based map. SSR map

GBS map

No. of progeny

119

424

No. of markers

120

1449

No. of linkage groups

21

19

Map size (cM)

1784

2424

Average distance between markers (cM)

14.87

1.67

Average genome size per cM (Mb/cM)

0.257

0.189

Genome coverage (%)

83.7

95.1

doi:10.1371/journal.pone.0149560.t001


8 / 19


Fig 3. Berry color segregation used for genetic map validation. (A) QTL for berry skin color is confirmed on Chr02 of the genetic map. Permutation tests were carried out to identify the 95% confidence threshold, and the significance threshold for the LOD score is presented as a horizontal red dashed line. (B) The additive effect (a, blue line) and dominant deviation (d, red line) are estimated across the whole genome. doi:10.1371/journal.pone.0149560.g003

Mbp and explained 26.0% of the phenotypic variance (Table 2). The additive effect was positive, indicating that higher MA/SS ratios came from ‘Seyval’, and the dominant effect was negligible. Since the QTL analysis was performed on data for a single year and single location, data transformation and multiple QTL detection methods and software programs were applied to confirm the results. The joint analysis of SS and MA showed evidence of two QTLs affecting both traits (Fig 4C). One QTL was detected at LG01 at the 49.94cM (5.0 Mbp), for which the additive effects indicated that the traits came from the grandparent 1, V. riparia (Table 2). This QTL explained around 7.9% and 1.0% of the phenotypic variance for MA and SS, respectively.


9 / 19


Table 2. Summary of QTL analyses in V. riparia × ‘Seyval’ F2 family. Trait

No. of individuals phenotyped

Linkage Group (LG)

Peak position (cM) a

Physical position of the nearest marker

LOD score

LOD interval (cM) b

LOD interval (physical location)

Additive effect c

Dominant effect

R2 d

Berry skin color

65

2

85.40

9,126,568

12.47

71.06– 90.85

6,360,259– 14,300,736

-0.50

0.47

89.92

Yeast assimilable nitrogen (YAN, mg/L)

63

7

100.44

18,819,915

6.48

94.87– 106.85

16,987,038– 20,565,550

-52.55

-38.91

22.81

Malic acid concentration (MA, g/L)

63

6

70.24

7,985,435

6.24

62.95– 75.23

7,343,698– 8,358,059

2.50

1.26

26.19

Total soluble solids content (SS, %w/w)

63

6

46.18

5,498,696

4.76

35.85– 54.71

4,258,443– 6,497,323

-2.16

0.0811

19.14

MA/SS ratio

63

6

32.10

3,857,140

5.93

27.07– 35.85

3,524,817– 4,258,443

0.17

0.00262

26.04

Joint analysis of MA and SS

63

1

49.94

5,061,702

6.55

49.10– 51.12

5,026,255– 5,123,842

MA -0.40

0.34

7.93

SS -0.14

0.20

1.02

Joint analysis of MA and SS

63

6

75.23

8,358,059

4.23

70.21– 80.83

7,985,435– 11,876,418

MA 0.60

0.28

18.37

SS 0.62

-0.24

19.43

a

Position of likelihood peak (highest LOD score).

b

LOD interval refers to 1.8-LOD support interval, except for Berry skin color, for which a 1.5-LOD support interval was used. Additive effect: A positive value means the higher value of the trait due to allele from grandparent2, ‘Seyval’. A negative value means the higher value of

c

the traits due to allele from grandparent1, V. riparia. The units depend on the traits. d

R2 (coefficient of determination): percentage of phenotypic variance explained by the QTL.

doi:10.1371/journal.pone.0149560.t002

Another QTL on LG06 at 75.23cM (8.3 Mbp) also showed high evidence for influencing both traits simultaneously; however, the additive positive effect on MA came from grandparent 2, ‘Seyval’, and the additive effect on SS came from V. riparia. This particular QTL was noteworthy since it explained around 18.3 and 19.4% of the phenotypic variance of MA and SS, respectively, and co-localized with the MA QTL identified in the CIM analysis. No significant dominance effect was observed for either trait. The interaction between QTL and trait additive effects was significant (Fig 4D). Thus for SS, both QTLs showed negative additive effects (-0.14 and -0.62, for LG01 and LG06, respectively), while for MA, the QTL on LG01 showed a negative effect and the QTL on LG06 positive (-0.40 and 0.60, respectively). The results from the causal model selection hypothesis tests showed that the values for Bayesian Information Criterion (BIC) and Akaike Information Criterion (AIC) did not agree in their most adequate model (S5 File). The AIC value suggested that the full model with nine parameters was best, and that the traits were affected by more than genetic factors. The BIC value suggested that considering SS as phenotype 1, the best model was a causal effect of SS on MA. When MA was considered as phenotype 1, the BIC value indicated the reactive model was best, confirming that MA was reacting to a causal effect of SS.

Discussion Over the last ten years, linkage map construction has been widely used in grapevine genetic research, with markers primarily being amplified fragment length polymorphisms (AFLPs),


10 / 19


Fig 4. QTL mapping for quantitative enological traits. (A) QTL mapping of yeast assimilable nitrogen (YAN, mg/L). (B) QTL mapping of malic acid concentration (MA, g/L). (C) The joint QTL mapping of MA (g/L) and total soluble solids content (SS, %w/w). Permutation tests were carried out to identify the 95% confidence threshold, and the significance thresholds of LOD score is presented as a horizontal red line in A-C. (D) The interaction between MA (g/L) and SS (%w/w) QTLs and trait additive effects. Red bars represent additive effects of MA and yellow bars represent additive effects of SS. A positive value means the higher value of the trait is due to an allele from grandparent 2, ‘Seyval’. A negative value means the higher value of the trait is due to an allele from grandparent 1, V. riparia. doi:10.1371/journal.pone.0149560.g004

SSRs and SNPs [27]. Although SSRs have been widely used in grapevine linkage map construction, the method is relatively time consuming and costly, typically resulting in low marker density. For example, 119 individuals from the mapping family used here were previously genotyped using 115 SSRs [29], which was much less than that required to capture all the informative recombination. That SSR map was unable to cover the whole genome and resulted in six gaps larger than 20 cM due to uneven distribution, which would result in information loss if causal alleles were located in the missing region. In contrast, GBS-based SNPs can provide high-density genetic maps and are conducive to high-throughput genotyping. The genetic map for this F2 population (Fig 2, Table 1) consisted of 1,449 markers 12.6-fold increase in marker density compared to the earlier SSR map. The maximum spacing between markers was reduced to 13.5 cM, and 15 out of 19 chromosomes have maximum gaps less than 10 cM, resulting in an average distance of 1.67 cM/marker. The GBS markers are almost evenly distributed across the genome, and cover 95% of the ‘PN40024’ reference genome. This high-resolution genetic map should better detect recombination events. Since linkage disequilibrium (LD) is low in Vitis due to its heterozygosity and diversity [11, 59], greater marker density increases the odds of finding markers for specific traits, or even to fine map candidate causal genes. Further, the GBS protocol is well established, publicly available, and can be scaled efficiently up to the current 384-plex barcoding system [35], at which the total genotyping cost is currently less than $15 per sample. Thus, GBS-SNP markers are a time- and cost-efficient method for high resolution genetic map construction and potentially for marker-assisted selection. However, current GBS pipelines have been designed for inbred germplasm with very low residual heterozygosity and do not provide methodology for


11 / 19


obtaining intercross markers, for which heterozygous materials are enriched and for which the marker phase is not known. The co-dominant intercross markers of an F2 mapping family enabled the capture of additional meiotic events relative to the F1 pseudo-testcross mapping approach. Furthermore, in an F1 family, only QTLs in heterozygous allelic states in the parents with large allele-effect differentials can be detected [60], which complicates the estimation of additive effects contributed by each parent. Thus, F2 families with at least 200 progeny genotyped with intercross markers have superior linkage map accuracy [61, 62] and allow the estimation of additive, and to some extent dominant, genetic effects contributed per parent [63]. There are two types of markers in an F2 mapping family: 1:2:1 segregation of co-dominant markers, which are more informative than 3:1 segregation of dominant markers. Here, we addressed several specific challenges associated with GBS analysis, particularly of a heterozygous family. First, retrieving useful GBS markers for map generation poses a challenge, since there is still a gap between the standard GBS variant calling pipeline TASSEL and commercial map construction software JoinMap. Also, heterozygous genotypes may be called due to sequencing errors or genome structure variance such as paralogs or tandem duplication. When imputation is difficult to apply on highly heterozygous species, pre-filtering and marker type identification and selection need to be carefully performed. The custom perl script (S1 File) provided in this paper identified markers with a co-dominant segregation (i.e. 1:2:1) in the progeny and verified each marker against the allelic state of the progenitor. To our knowledge, this is the first publication using intercross SNP markers for map construction in an F2 progeny in grapevine. The perl script can be used for marker development in other highly heterozygous species, such as apple, poplar and willow. To validate the marker development and map construction strategy, the trait of berry color was subjected to QTL analysis (Fig 3). Berry color was selected for map validation for two reasons: (1) it is a highly heritable binary trait that is easy to phenotype; and (2) the genetic basis is well studied. Specifically, the VvmybA1 gene (Chr02: 14,179,266–14,180,746) regulates anthocyanin pigment biosynthesis in grapes [64], and a retrotransposon-induced mutation in VvmybA1 has been identified as the molecular basis of white cultivars of V. vinifera [58]. The QTL detected at 14.2 Mbp on Chr02 aligned perfectly with VvmybA1, indicating our approach was effective, even with a small sample size of 65 phenotyped progeny. The GBS genetic map was also compared to a physical map based on the reference genome V. vinifera ‘PN40024’ (S3 File). The majority of co-linearity indicated that marker order is well conserved even across the genome segments of diverse Vitis species represented in the parents, including V. vinifera, V. rupestris, V. aestivalis var lincecumii and V. riparia. These markers can be used as anchors to improve the ‘PN40024’ reference genome assembly by filling existing gaps or incorporating random contigs. However, there are some chromosomal regions that show weaker correlation between genetic and physical maps, e.g. the upper end of Chr16. These regions might indicate species-specific genome structure variation, such as chromosome rearrangement, transposable elements and tandem duplication. V. vinifera is the only species in the Vitis complex with a whole genome sequence already published, and the de novo genome assembly for other Vitis species can provide insight into unique characteristics arising from adaption to local conditions. Thus, this interspecific genetic map can be viewed as an early step toward a grapevine pan-genome, as pursued in maize [64]. Once the genetic map was generated and validated, three berry traits related to winegrape quality were chosen for QTL analysis. YAN–or the sum of nitrogen contained in primary amino acids and ammonium–is the nitrogenous component of grapes that can be utilized by yeast during alcoholic fermentation. Proper nitrogen concentrations are of interest to winemakers since nitrogen deficiency is associated with sluggish or stuck fermentations, increased


12 / 19


production of off-aroma compounds like H2S, and decreased formation of desirable odorants like esters [1, 65]. Conversely, excess YAN can cause several problems, most notably formation of carcinogenic ethyl carbamate. Typically, ammonium is accumulated first, and decreases during ripening with corresponding increases in primary amino nitrogen [1]. Total YAN changes relatively little during berry ripening–across a wide range of cultivars and sites, YAN concentrations at commercial harvest averaged only 25% higher and were well correlated with values measured up to 5 weeks prior [66]. However, genotype can have a profound effect on YAN–up to a three-fold variation among vinifera cultivars (75 mg/L vs. 220 mg/L) was observed in a survey of cultivars in one growing region [66], and much larger variations were observed among interspecific hybrids [18]. We observed a YAN range in our mapping family (100–600 mg/L) greater than this previous survey, further confirming the importance of genetics in YAN. Currently there is little information on the genetics of YAN and this, to our knowledge, is the first YAN associated QTL in grapevine. The QTL results in this study revealed one locus associated with YAN on LG07 (LOD = 6.48) from 16,987,038 to 20,565,550 bp, within the 1.8-LOD support interval (Table 2). Thirteen genes relevant to nitrogen metabolism were identified within the interval and can be used to further explore the trait. Because berries appear to accumulate ammonium, which is subsequently metabolized to primary amino acids, the presence of ammonium transporter genes in this list (AMT2) is of particular interest. The altered function of these genes could limit either N uptake by the vines or YAN accumulation in the berries. The estimated additive effect of the QTL indicated the V. riparia grandparent contained a higher concentration of YAN than ‘Seyval’ and may provide a novel tool for research into the nature of nitrogen content in grape berries. Both MA and SS are critical parameters for evaluating fruit maturity, where the primary component of SS in mature fruit is hexose sugars (fructose, glucose). The key biochemical pathways associated with MA and hexose sugar metabolism are well-established [67]. Pre-veraison, sucrose imported into grape berries via the phloem is used as the primary metabolic substrate. Concurrently, a portion of malate from the citric acid cycle is transported and stored in vacuoles, where it serves as a key contributor (along with tartaric acid) to the high TA of unripe grapes [68]. During veraison and ripening, grapes accumulate sugars, and shift to using MA as an energy source [14]. Thus, a key change associated with veraison is an increase in SS (caused primarily by inversion of imported sucrose and accumulation of resulting hexose sugars) coinciding with a reduction in TA (caused primarily because of MA respiration). Grape crosses with non-domesticated Vitis in their background, e.g. V. riparia, can have unacceptably high TA values and MA concentrations (>10 g/L) even at normal harvest SS values (>20 Brix) [5]. The family used in this study shows this problem with over 50% of individuals having MA > 10 g/L (S4 File). Because MA and SS are co-regulated, QTL analyses of individual traits were complemented by two other approaches 1) QTL analysis of the ratio, MA/SS, and 2) joint (simultaneous multi-trait) analysis of MA and SS. Both techniques have advantages over the trait-by-trait analysis, the most relevant being the increase in power to identify and accurately locate QTLs [69–72]. However, both approaches also provide additional information that is not easily obtained when traits are analyzed separately, such as the power to identify loci involved in the regulation at specific metabolic branch-points and infer interaction among genetic components for the exhibition of the traits. The analysis of two traits as a ratio creates some difficulties, because when two traits are identically normally distributed, their ratio follows a Cauchy distribution as opposed to the normal distribution expected by most algorithms [71]. However, the mapping of these ratiotraits is relevant because it allows researchers to test and infer the interaction between traits, exemplified here by the interaction of compounds that have relevance for harvest decision-


13 / 19


making in the field and fermentation must manipulations in the winery. From a genetics perspective, the analysis of ratio-traits allows the potential opportunity to integrate and discover independent, closely-linked, and pleiotropic loci [73] when the separated traits exhibit a significant positive or negative correlation. For example, MA/SS ratio would be less sensitive than the individual traits to changes in berry size. In our study, the analysis of the MA/SS ratio showed a QTL on LG06 distal from the QTLs identified for SS and MA when analyzed separately (Table 2, S6 File). The LOD-based evidence (4.77–6.24) and amount of variance explained (19–26%) was comparable for the traits analyzed individually or as a ratio. The interpretation of the additive and dominance effects suggested that the higher value for the ratio could be due to the genetic contribution from ‘Seyval’; however, since this ratio can be interpreted as a biochemical interaction, the interpretation may require additional inference to suggest what is being genetically inherited. The joint analysis of multiple traits includes all traits of interest simultaneously in a single model. This approach can provide insights into fundamental genetic mechanisms underlying trait relationships such as pleiotropy and close linkage, genotype-by-environment (G × E) interactions, and the possible trade-off given for negatively correlated traits [70]. This joint analysis approach generated a list of 14 candidate genes within the QTL region. The joint analysis showed a QTL detected on Chr06 with 1.8-LOD support interval from 70.2–80.8 cM (Table 2). The corresponding physical interval, Chr06:7,985,435–11,876,418, covered the region of grapevine orthologs of the aluminum activated malate transporter, ALMT1. This gene corresponds to the Ma1 and Ma2 loci in apple, which are malic acid transporters controlling the acidity level in apple fruits [74]. Our observation aligns with a QTL for MA at 74.1 cM on Chr06, reported by another group using a Vitis spp. mapping family [20]. The QTL explained 17% of the variance for MA in one year of a three year study. This group did not implicate ALMT1 as a candidate gene in Chr06, although they did observe a more minor QTL on Chr18, which they suggested could be related to a different ALMT. Interestingly, significant QTL were not detected for other genes expected to control MA concentrations. For example, MA respiration rate, the rate at which grape berries metabolize MA after veraison, is dependent on NAD dependent malate enzyme [68], but no QTL was detected at its genetic location. QTL analyses of MA are complicated because high MA at harvest could arise from either high MA accumulation pre-veraison or slow MA respiration, and future studies could likely benefit from having multiple time points, i.e. pre- and post-veraison. The analysis of multiple traits simultaneously implies advantages when causal relationships between the studied traits and genes or among traits exist; however, when the trait is affected by a gene through a transitional trait, the advantages almost disappear [72]. For such a reason the testing of the causal relationships (correlation, pleiotropy, or close linkage) between or among traits must be addressed to yield outputs that can be interpreted biologically. Also, the addition of more traits to the joint QTL analysis does not result in more detection power, but may give rise to spurious signals since additional parameters will be fitted in the model [75]. However, to break unfavorable linkages of genes involved in the exhibition of economically important traits, the separation of closely linked loci is needed to eliminate pleiotropic negative effects [75]. In this study, the test for causal relationships between SS and MA was pursued through causal model selection hypothesis tests [56]. The results from CMST showed that a causal relationship clearly exists between MA and SS with a common genetic component (QTL), as the model for independence was not significant in any of the tests. AIC values were minimized for the full model, indicating SS and MA share more than one common genetic component. While, BIC values were minimized for models showing a casual effect of a QTL controlling SS also affecting MA. The discrepancy between AIC and BIC is not a concern for the


14 / 19


interpretation of the results in the light of the empirical experience for these traits. Therefore, the results are compatible with the empirical experience, in which the balance between MA and SS is intertwined and influenced by several other factors, such as environment (particularly temperature), field management (irrigation, fertilization), and biology (additional biochemical interactions). Hence, genetic research, breeding, and probably biochemistry approaches, should consider MA and SS simultaneously, as well as the possible interactions with nongenetic factors.

Conclusions The results presented here confirmed the GBS mapping strategy for an expanded F2 mapping family in grapevine. The protocol presented here filters heterozygous markers and retrieves high quality intercross markers which are phased during linkage map development and allows QTL analysis using an F2 genetic design, which is not common for grapevine genetic research. In addition, the protocols for calling intercross GBS markers could be used for other heterozygous species. The linkage map provided qualitative and quantitative improvements over a previous SSR map for marker density and genome coverage while maintaining co-linearity with the reference genome. The availability of the linkage map facilitated the genetic study of winegrape quality traits, including a well-characterized binary trait (berry color), more complicated traits (malic acid, soluble solids) and a neglected trait (yeast assimilable nitrogen). Enhanced marker resolution provided the opportunity to conduct joint analysis of malic acid and soluble solids and results indicated these compounds are intertwined and should be considered jointly in future genetic studies. The resources generated here can further guide the study and manipulation of fruit harvest, must processing and wine fermentation, and may also contribute to decision making for breeding and cultivar development of grapevines. The novel GBS analysis methods presented here could have broad impact for F2 mapping in diverse heterozygous species.

Supporting Information S1 File. Perl code for obtaining intercross markers. Perl code to identify markers from TASSEL hapmap file that are (1) heterozygous in the F1 progenitor and (2) show co-dominant 1:2:1 segregation among the F2 progeny by a chi-squared (χ2) goodness-of-fit test at α0.01. (PL) S2 File. GBS-based linkage map statistics. Summary statistics of the V. riparia × ‘Seyval’ F2 GBS-based linkage map. (DOCX) S3 File. Comparison between GBS genetic map and physical map. X-axes represent physical coordinates in the reference genome V. vinifera ‘PN40024’ version 12X.v2 (Mb). Y-axes represent genetic coordinates (cM). (PDF) S4 File. Distribution of traits in the V. riparia × ‘Seyval’ F2 family. (A) Sixty five F2 progeny were measured for berry skin color. White (nonpigmented) is coded as 0 and black (pigmented) is coded as 1. (B-D) Distribution frequency for quantitative enological traits in 63 F2 progeny. (PDF) S5 File. Results from the causal model selection test. Log-likelihood (LogLik), Akaike Information Criterion (AIC) and Bayesian Information Criterion (BIC) evaluated for each of the


15 / 19


causal models considered in CMST. In the column “Test”, a simple directed acyclic diagram is shown as a representation of the test performed. In this particular case, a genetic componentQTL (Q) influences either of the traits (MA or SS). The arrows represent the direction of the influence, which can go from Q to either trait or from trait to trait. Influence of trait upon Q is not considered in this approach. The values associated with the most likely models are highlighted in bold. (DOCX) S6 File. QTL mapping for berry quality traits. QTL mapping for (A) total soluble solids content (SS, %w/w) and (B) the ratio of malic acid concentration (MA, g/L) to total soluble solids content (SS, %w/w). Permutation tests were carried out to identify 95% confidence thresholds, and the significance threshold of LOD score is presented as a horizontal red dashed line. (PDF) S7 File. SNP marker sequences. The sequence of markers in the final map. (FASTA)

Acknowledgments The authors thank Kalley Besler and Mani Awale for fruit collection and preprocessing and the South Dakota State University Agricultural Experiment Station NE Hansen Research Farm where the grapevine population is maintained.

Author Contributions Conceived and designed the experiments: AYF JJL QS AKM JPL BIR LCD. Performed the experiments: SY JFR QS DCM GLS AYF BIR LCD. Analyzed the data: SY JFR QS DCM GLS AYF BIR LCD. Wrote the paper: SY JFR.

References 1.

Bell S- J, Henschke PA. Implications of nitrogen nutrition for grapes, fermentation and wine. Aust J Grape Wine Res. 2005; 11(3): 242–95. doi: 10.1111/j.1755-0238.2005.tb00028.x

2.

Mahanil S, Ramming D, Cadle-Davidson M, Owens C, Garris A, Myles S, et al. Development of marker sets useful in the early selection of Ren4 powdery mildew resistance and seedlessness for table and raisin grape breeding. Theoret Appl Genet. 2012; 124(1): 23–33. doi: 10.1007/s00122-011-1684-7

3.

Sacks GL, Acree TE, M.T. K, Vanden Heuvel JE, Wilcox WF, editors. Tell me about your childhood”– the role of the vineyard in determining wine flavour chemistry. 15th Australian wine industry technical conference; 2013; Sydney, New South Wales, Australia.

4.

Ugliano M, Henschke P. Yeasts and wine flavour. In: Moreno-Arribas MV, Polo MC, editors. Wine chemistry and biochemistry: Springer New York; 2009. p. 313–92.

5.

Reisch B, Owens C, Cousins P. Grape. In: Badenes ML, Byrne DH, editors. Fruit breeding. Handbook of plant breeding. Springer US; 2012. p. 225–262.

6.

Young PR, Vivier MA. Genetics and genomic approaches to improve grape quality for winemaking. In: Reynolds AG, editor. Managing wine quality: Woodhead Publishing; 2010. p. 316–364.

7.

Di Gaspero G, Cattonaro F. Application of genomics to grapevine improvement. Aust J Grape Wine Res. 2010; 16: 122–130. doi: 10.1111/j.1755-0238.2009.00072.x

8.

Sawler J, Reisch B, Aradhya MK, Prins B, Zhong GY, Schwaninger H, et al. Genomics assisted ancestry deconvolution in grape. PLoS One. 2013; 8(11): e80791. doi: 10.1371/journal.pone.0080791 PMID: 24244717

9.

Jackson RS. Wine science: Principles and applications. 4th ed. San Diego: Academic Press; 2014.

10.

Myles S. Improving fruit and wine: what does genomics have to offer? Trends Genet. 2013; 29(4): 190–196. doi: 10.1016/j.tig.2013.01.006 PMID: 23428114


16 / 19


11.

Myles S, Boyko AR, Owens CL, Brown PJ, Grassi F, Aradhya MK, et al. Genetic structure and domestication history of the grape. Proc Natl Acad Sci USA. 2011; 108(9): 3530–3535. doi: 10.1073/pnas. 1009363108 PMID: 21245334

12.

Sun Q, Gates MJ, Lavin EH, Acree TE, Sacks GL. Comparison of odor-active compounds in grapes and wines from Vitis vinifera and non-foxy American grape species. J Agric Food Chem. 2011; 59(19): 10657–10664. doi: 10.1021/jf2026204 PMID: 21879766

13.

Boulton R, Singleton V, Bisson L, Kunkee R. Preparation of musts and juice. Principles and practices of winemaking: Springer US; 1999. p. 65–101.

14.

Volschenk1a H, Van Vuuren H, Viljoen-Bloom M. Malic acid in wine: origin, function and metabolism during vinification. S Afr J Enol Vitic. 2006; 27(2): 123–136.

15.

Kliewer WM. Concentration of Tartrates, Malates, Glucose and Fructose in the Fruits of the Genus Vitis. Am J Enol Vitic. 1967; 18(2): 87–96.

16.

Jackson D. Climate, monographs in cool climate viticulture-2. Daphne Brasell NZ. 2001.

17.

Brice C, Sanchez I, Bigey F, Legras JL, Blondin B. A genetic approach of wine yeast fermentation capacity in nitrogen-starvation reveals the key role of nitrogen signaling. BMC Genomics. 2014; 15:495. doi: 10.1186/1471-2164-15-495 PMID: 24947828

18.

Stewart ACH. Nitrogen composition of interspecific hybrid and Vitis vinifera wine grapes from the Eastern United States. Ph.D. Dissertation. Purdue University; 2013. Available: http://docs.lib.purdue.edu/ dissertations/AAI3592130/.

19.

Song S, del Mar Hernández M, Provedo I, Menéndez C. Segregation and associations of enological and agronomic traits in Graciano × Tempranillo wine grape progeny (Vitis vinifera L.). Euphytica. 2014; 195(2): 259–277. doi: 10.1007/s10681-013-0994-z

20.

Chen J, Wang N, Fang L-C, Liang Z-C, Li S-H, Wu B-H. Construction of a high-density genetic map and QTLs mapping for sugars and acids in grape berries. BMC Plant Biol. 2015; 15: 28. doi: 10.1186/ s12870-015-0428-2 PMID: 25644551

21.

Viana AP, Riaz S, Walker MA. Genetic dissection of agronomic traits within a segregating population of breeding table grapes. Genet Mol Res. 2013; 12(2): 951–964. doi: 10.4238/2013.April.2.11 PMID: 23613241

22.

Huang Z, Ough CS. Effect of Vineyard Locations, Varieties, and Rootstocks on the Juice Amino Acid Composition of Several CultivarsAm J Enol ViticAm J Enol Vitic. 1989; 40(2): 135–139.

23.

Grattapaglia D, Bertolucci FL, Sederoff RR. Genetic mapping of QTLs controlling vegetative propagation in Eucalyptus grandis and E. urophylla using a pseudo-testcross strategy and RAPD markers. Theor Appl Genet. 1995; 90(7–8): 933–947. doi: 10.1007/BF00222906 PMID: 24173047

24.

Adam-Blondon AF, Roux C, Claux D, Butterlin G, Merdinoglu D, This P. Mapping 245 SSR markers on the Vitis vinifera genome: a tool for grape genetics. Theor Appl Genet. 2004; 109(5): 1017–1027. doi: 10.1007/s00122-004-1704-y PMID: 15184982

25.

Barba P, Cadle-Davidson L, Harriman J, Glaubitz JC, Brooks S, Hyma K, et al. Grapevine powdery mildew resistance and susceptibility loci identified on a high-resolution SNP map. Theor Appl Genet. 2014; 127(1): 73–84. doi: 10.1007/s00122-013-2202-x PMID: 24072208

26.

Doligez A, Adam-Blondon AF, Cipriani G, Di Gaspero G, Laucou V, Merdinoglu D, et al. An integrated SSR map of grapevine based on five mapping populations. Theor Appl Genet. 2006; 113(3): 369–82. doi: 10.1007/s00122-006-0295-1 PMID: 16799809

27.

Wang N, Fang LC, Xin HP, Wang LJ, Li SH. Construction of a high-density genetic map for grape using next generation restriction-site associated DNA sequencing. BMC Plant Biol. 2012; 12:148. doi: 10. 1186/1471-2229-12-148 PMID: 22908993

28.

Fennell A, Mathiason K, Luby J. Genetic segregation for indicators of photoperiod control of dormancy induction in Vitis species. Acta Hortic. 689. 2005; p. 533–540. doi: 10.17660/ActaHortic.2005.689.66

29.

Garris A, Clark L, Owens C, McKay S, Luby J, Mathiason K, et al. Mapping of photoperiod-induced growth cessation in the wild grape Vitis riparia. J Am Soc Hortic Sci. 2009; 134(2): 261–272.

30.

Fennell AY, Schlauch KA, Gouthu S, Deluc LG, Khadka V, Sreekantan L, et al. Short day transcriptomic programming during induction of dormancy in grapevine. Front. Plant Sci. 2015; 6:834. doi: 10.3389/ fpls.2015.00834

31.

Mathiason K, He D, Grimplet J, Venkateswari J, Galbraith DW, Or E, et al. Transcript profiling in Vitis riparia during chilling requirement fulfillment reveals coordination of gene expression patterns with optimized bud break. Funct Integr Genomics. 2009; 9(1): 81–96. doi: 10.1007/s10142-008-0090-y PMID: 18633655

32.

Sreekantan L, Mathiason K, Grimplet J, Schlauch K, Dickerson JA, Fennell AY. Differential floral development and gene expression in grapevines during long and short photoperiods suggests a role for floral


17 / 19


genes in dormancy transitioning. Plant Mol Biol. 2010; 73(1–2): 191–205. doi: 10.1007/s11103-0109611-x PMID: 20151315 33.

Glaubitz JC, Casstevens TM, Lu F, Harriman J, Elshire RJ, Sun Q, et al. TASSEL-GBS: A High Capacity Genotyping by Sequencing Analysis Pipeline. PLoS One. 2014; 9(2): e90346. doi: 10.1371/journal. pone.0090346 PMID: 24587335

34.

Swarts K, Li HH, Navarro JAR, An D, Romay MC, Hearne S, et al. Novel Methods to Optimize Genotypic Imputation for Low-Coverage, Next-Generation Sequence Data in Crop Plants. Plant Genome. 2014; 7(3). doi: 10.3835/plantgenome2014.05.0023

35.

Elshire RJ, Glaubitz JC, Sun Q, Poland JA, Kawamoto K, Buckler ES, et al. A Robust, Simple Genotyping-by-Sequencing (GBS) Approach for High Diversity Species. PLoS One. 2011; 6(5): e19379. doi: 10.1371/journal.pone.0019379 PMID: 21573248

36.

Azuma A, Kobayashi S, Mitani N, Shiraishi M, Yamada M, Ueno T, et al. Genomic and genetic analysis of Myb-related genes that regulate anthocyanin biosynthesis in grape berry skin. Theor Appl Genet. 2008; 117(6): 1009–1019. doi: 10.1007/s00122-008-0840-1 PMID: 18651125

37.

Cardoso S, Lau W, Dias JE, Fevereiro P, Maniatis N. A Candidate-Gene Association Study for Berry Colour and Anthocyanin Content in Vitis vinifera L. PLoS One Sep 28. 2012; 7(9): e46021. doi: 10. 1371/journal.pone.0046021 PMID: 23029369

38.

Salmaso M, Malacarne G, Troggio M, Faes G, Stefanini M, Grando MS, et al. A grapevine (Vitis vinifera L.) genetic map integrating the position of 139 expressed genes. Theor Appl Genet. 2008; 116(8): 1129–1143. doi: 10.1007/s00122-008-0741-3 PMID: 18347774

39.

Castellari M, Versari A, Spinabelli U, Galassi S, Amati A. An improved HPLC method for the analysis of organic acids, carbohydrates, and alcohols in grape musts and wines. J Liq Chromatogr Relat Technol. 2000; 23(13): 2047–2056. doi: 10.1081/Jlc-100100472

40.

Jaillon O, Aury JM, Noel B, Policriti A, Clepet C, Casagrande A, et al. The grapevine genome sequence suggests ancestral hexaploidization in major angiosperm phyla. Nature. 2007; 449(7161): 463–467. doi: 10.1038/nature06148 PMID: 17721507

41.

Li H, Durbin R. Fast and accurate short read alignment with Burrows-Wheeler transform. Bioinformatics. 2009; 25(14):1754–1760. doi: 10.1093/bioinformatics/btp324 PMID: 19451168

42.

Danecek P, Auton A, Abecasis G, Albers CA, Banks E, DePristo MA, et al. The variant call format and VCFtools. Bioinformatics. 2011; 27(15): 2156–8. doi: 10.1093/bioinformatics/btr330 PMID: 21653522

43.

Bradbury PJ, Zhang Z, Kroon DE, Casstevens TM, Ramdoss Y, Buckler ES. TASSEL: software for association mapping of complex traits in diverse samples. Bioinformatics. 2007; 23(19): 2633–2635. doi: 10.1093/bioinformatics/btm308 PMID: 17586829

44.

Van Ooijen J. JoinMap1 4. Software for the calculation of genetic linkage maps in experimental populations. Version 4.1. Wageningen, The Netherlands: Kyazma; 2006.

45.

Wu R, Ma C-X, Painter I, Zeng Z-B. Simultaneous maximum likelihood estimation of linkage and linkage phases in outcrossing species. Theoretical Popul Biol. 2002; 61(3): 349–363. doi: 10.1006/tpbi.2002. 1577

46.

Van Ooijen J. Multipoint maximum likelihood mapping in a full-sib family of an outbreeding species. Genet Res. 2011; 93(5): 343–349. doi: 10.1017/S0016672311000279

47.

Broman KW, Wu H, Sen S, Churchill GA. R/qtl: QTL mapping in experimental crosses. Bioinformatics. 2003; 19(7): 889–890. doi: 10.1093/Bioinformatics/Btg112 PMID: 12724300

48.

R Core Team. R: A language and environment for statistical computing. Version 3.1.2. Vienna, Austria: R Foundation for statistical computing; 2014.

49.

Boer MP, Malosetti M, Welham SJ, Thissen JTNM. Part 2: Statistical genetics and QTL estimation. In: Payne RW, Harding SA, Murray DA, Soutar DM, Baird DB, Glaser AI, et al., editors. The guide to GenStat release 17. Hemel Hempstead, UK: VSN International; 2014.

50.

VSN International. GenStat for Windows 17th edition. Hemel Hempstead, UK: VSN International; 2014.

51.

Broman KW, Sen S. A guide to QTL mapping with R/qtl: Springer; 2009.

52.

Li J, Ji L. Adjusting multiple testing in multilocus analyses using the eigenvalues of a correlation matrix. Heredity. 2005; 95(3): 221–7. doi: 10.1038/Sj.Hdy.6800717 PMID: 16077740

53.

Malosetti M, Voltas J, Romagosa I, Ullrich SE, van Eeuwijk FA. Mixed models including environmental covariables for studying QTL by environment interaction. Euphytica. 2004; 137(1):139–145. doi: 10. 1023/B:Euph.0000040511.46388.Ef

54.

Boer MP, Wright D, Feng L, Podlich DW, Luo L, Cooper M, et al. A mixed-model quantitative trait loci (QTL) analysis for multiple-environment trial data using environmental covariables for QTL-by-


18 / 19


environment interactions, with an example in maize. Genetics. 2007; 177(3): 1801–1813. doi: 10.1534/ genetics.107.071068 PMID: 17947443 55.

Lynch M, Walsh B. Genetics and analysis of quantitative traits. Sunderland, Mass. Sinauer; 1998.

56.

Neto EC, Broman AT, Keller MP, Attie AD, Zhang B, Zhu J, et al. Modeling causality for pairs of phenotypes in system genetics. Genetics. 2013; 193(3): 1003–1013. doi: 10.1534/Genetics.112.147124/-/ Dc1 PMID: 23288936

57.

Neto EC, Yandell BS. qtlhot: Inference for QTL Hotspots. 2013. R package version 0.9.0. Available: http://CRAN.R-project.org/package=qtlhot.

58.

Kobayashi S, Goto-Yamamoto N, Hirochika H. Retrotransposon-induced mutations in grape skin color. Science. 2004; 304(5673): 982. doi: 10.1126/science.1095011 PMID: 15143274

59.

Myles S, Chia J-M, Hurwitz B, Simon C, Zhong GY, Buckler E, et al. Rapid genomic characterization of the genus Vitis. PLoS One. 2010 January 13, 2010; 5(1): e8219. doi: 10.1371/journal.pone.0008219 PMID: 20084295

60.

Plomion C, Durel C. Estimation of the average effects of specific alleles detected by the pseudo-testcross QTL mapping strategy. Genet Sel Evol. 1996; 28(3): 1–13. doi: 10.1186/1297-9686-28-3-223

61.

Lorieux M, Perrier X, Goffinet B, Lanaud C, de León DG. Maximum-likelihood models for mapping genetic markers showing segregation distortion. 2. F2 populations. Theor Appl Genet. 1995; 90(1): 81– 89. doi: 10.1007/BF00220999 PMID: 24173787

62.

Staub JE, Serquen FC, Gupta M. Genetic markers, map construction, and their application in plant breeding. HortScience. 1996; 31(5): 729–741.

63.

Van Ooijen J, Jansen J. Genetic mapping in experimental populations. Cambridge: Cambridge University Press; 2013.

64.

Morgante M, De Paoli E, Radovic S. Transposable elements and the plant pan-genomes. Curr Opin Plant Biol. 2007; 10(2):149–155. doi: 10.1016/j.pbi.2007.02.001 PMID: 17300983

65.

Ugliano M, Kolouchova R, Henschke PA. Occurrence of hydrogen sulfide in wine and in fermentation: influence of yeast strain and supplementation of yeast available nitrogen. J Ind Microbiol Biotechnol. 2011; 38(3): 423–429. doi: 10.1007/s10295-010-0786-6 PMID: 20668912

66.

Nisbet MA, Martinson TE, Mansfield AK. Accumulation and Prediction of Yeast Assimilable Nitrogen in New York Winegrape Cultivars. Am J Enol Vitic. 2014; 65(3): 325–332. doi: 10.5344/ajev.2014.13130

67.

Kanellis A, Roubelakis-Angelakis K. Grape. Biochemistry of fruit ripening: Springer; 1993. p. 189–234.

68.

Sweetman C, Deluc LG, Cramer GR, Ford CM, Soole KL. Regulation of malate metabolism in grape berry and other developing fruits. Phytochemistry. 2009; 70(11–12): 1329–1344. doi: 10.1016/j. phytochem.2009.08.006 PMID: 19762054

69.

Banerjee S, Yandell BS, Yi NJ. Bayesian quantitative trait loci mapping for multiple traits. Genetics. 2008; 179(4): 2275–2289. doi: 10.1534/Genetics.108.088427 PMID: 18689903

70.

Xu CW, Li ZK, Xu SZ. Joint mapping of quantitative trait loci for multiple binary characters. Genetics. 2005; 169(2): 1045–1059. doi: 10.1534/Genetics.103.019406 PMID: 15489509

71.

Yang RQ, Li JH, Xu SZ. Mapping quantitative trait loci for traits defined as ratios. Genetica. 2008; 132 (3): 323–329. doi: 10.1007/S10709-007-9175-0 PMID: 17671840

72.

Zhu W, Zhang H. Why do we test multiple traits in genetic association studies? J Korean Stat Soc. 2009; 38(1): 1–10. doi: 10.1016/j.jkss.2008.10.006 PMID: 19655045

73.

Yang RQ, Jin TB, Li WB. Mapping genome-wide QTL of ratio traits with Bayesian shrinkage analysis for its component traits. Genetica. 2010; 138(8): 853–860. doi: 10.1007/S10709-010-9468-6 PMID: 20556635

74.

Khan SA, Beekwilder J, Schaart JG, Mumm R, Soriano JM, Jacobsen E, et al. Differences in acidity of apples are probably mainly caused by a malic acid transporter gene on LG16. Tree Genet Genomes. 2013; 9(2): 475–487. doi: 10.1007/S11295-012-0571-Y

75.

Jiang C, Zeng ZB. Multiple trait analysis of genetic mapping for quantitative trait loci. Genetics. 1995; 140(3): 1111–1127. PMID: 7672582


19 / 19

Next Generation Mapping of Enological Traits in ... - Semantic Scholar

Next Generation Mapping of Enological Traits in ... - Semantic Scholar

Suggest Documents

Next-Generation Mapping of Complex Traits with ... - Semantic Scholar

Next-Generation Sequencing Approach in ... - Semantic Scholar

Next-generation sequencing in neuropathologic ... - Semantic Scholar

Next-generation sequencing in neuropathologic ... - Semantic Scholar

Uses of Next-Generation Sequencing ... - Semantic Scholar

Evaluation of next generation mtGenome ... - Semantic Scholar

Clinical applications of next generation ... - Semantic Scholar

Next Generation Semantic Web Applications - Semantic Scholar

Next generation sequencing of exceptional ... - Semantic Scholar

Next-generation sequencing of mitochondrial ... - Semantic Scholar

Evaluation of Next Generation Networks - Semantic Scholar

Association Mapping of Germination Traits in ... - Semantic Scholar

Targeted Next Generation Sequencing reveals ... - Semantic Scholar

Aspirin for the next generation - Semantic Scholar

Metabolic autopsy with next generation ... - Semantic Scholar

Next-Generation Optical Access Networks - Semantic Scholar

Next Generation Application Integration - Semantic Scholar

Next Generation Hypermedia Authoring Systems - Semantic Scholar

Next-generation sequencing-based molecular ... - Semantic Scholar

Next generation genome sequencing reveals ... - Semantic Scholar

Building Next-Generation Converged Networks - Semantic Scholar

convergence & next generation networks - Semantic Scholar

Tumor heterogeneity: next-generation sequencing ... - Semantic Scholar

Evolution for the Next Generation - Semantic Scholar