Skip to main content
Log in

Linkage mapping and QTL analysis of pecan (Carya illinoinensis) full-siblings using genotyping-by-sequencing

  • Original Article
  • Published:
Tree Genetics & Genomes Aims and scope Submit manuscript

Abstract

Pecan (Carya illinoinensis) is a nut-producing tree native to North America grown for its culinary, ornamental, and lumber characteristics. A newly developed chromosome-scale reference genome for pecan was used to create genotyping-by-sequencing (GBS)-based high-density genetic linkage maps of 151 full-sibling progeny of ‘Elliott’ and ‘VC1-68’. These maps incorporate 6142 SNPs segregating in a testcross pattern into 32 linkage groups representing the 16 chromosomes of pecan across the two parents. The average distance between markers was 0.46 cM and the two maps totaled 1376.4 cM and 1463.1 cM for ‘Elliott’ and ‘VC1-68’, respectively. These markers, plus an additional 1096 intercross markers, were used to create a 1557.8-cM pecan consensus genetic linkage map. Quantitative trait locus (QTL) analyses revealed 1 major and 2 minor effect QTL for budbreak and 1 minor effect QTL for pecan scab susceptibility. The major effect locus inherited from ‘VC1-68’ explained up to 30% of the variation in budbreak and appeared across 3 years of observations. This QTL is syntenic to recently identified major effect QTL for budbreak in English walnut. The techniques reported herein, and the resulting genetic maps will facilitate future discoveries of valuable fruiting trait loci as this pecan population completes the transition to sexual maturity. In addition, the major budbreak QTL coincident in pecan and English walnut may represent a conserved mechanism effecting budbreak across the Juglandaceae providing critical knowledge for future investigations of variability in phenology.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Fig. 1
Fig. 2
Fig. 3
Fig. 4

Similar content being viewed by others

References

Download references

Acknowledgments

The authors wish to thank Ms. Julie McCollum and Ms. Natalie Patterson for help in pecan DNA extraction and Illumina template preparation and Ms. Rory Tucker for her assistance evaluating SSRs. Illumina sequencing was provided by Texas A&M AgriLife Research Genomics and Bioinformatics Services. We also thank Linwood Nursery for providing access to their proprietary cultivar ‘VC1-68’ for the creation of this population. Questions regarding living inventories should be directed to the USDA-ARS Pecan Breeding and Genetics program.

Funding

This work was supported by the United States Department of Agriculture Agricultural Research Service (USDA-ARS) CRIS project 6202-21000-036-00D (Management and Characterization of Pecan Genetics Resources and Related Wild Populations), USDA-ARS CRIS project 6202-21000-035-00D (Pecan Improvement Through Breeding and Genetics), Specific Cooperative Agreement 58-6202-1-201 (Developing Molecular Markers for Carya), Specific Cooperative Agreement 58-3091-5-031 (Genomic Markers for Carya), USDA Hatch funds, and USDA-SCRI 58-6042-6009 (Coordinated Development of Genetic Tools for Pecan).

Author information

Authors and Affiliations

Authors

Contributions

LJ Grauke conceived the population’s creation, long-term care, and contributed expertise in monitoring phenotypes. Keith Kubenka contributed to the maintenance of the population as well as phenotypic evaluation. Erin Ruhlman and Nolan Bentley collected, extracted, and genotyped samples with SSRs including novel SSRs they developed. Robert Klein contributed access to the ABI3130xl and assistance with genotyping the population with SSRs. Robert Klein and Nolan Bentley contributed to GBS template DNA preparation. Xinwang Wang contributed the budbreak phenotype observations in 2018 and 2019. Patricia Klein analyzed the sequencing data and generated the SNP calls. Nolan Bentley contributed multiple phenotype evaluations, wrote the R scripts, and performed downstream analyses of the SNP and phenotypic data. Nolan Bentley wrote the manuscript with revisions and contributions to the interpretation from Patricia Klein, Robert Klein, and LJ Grauke.

Corresponding author

Correspondence to Patricia Klein.

Ethics declarations

Conflict of interest

The authors declare that they have no competing interests.

Data archiving statement

The sequences generated in this study have been submitted to the NCBI Sequence Read Archive (SRA) (http://www.ncbi.nlm.nih.gov/sra) under the BioProject number PRJNA602509. The reference sequence assembly used in this study (Oaxaca-v1.0, HudsonAlpha data unpublished) can be accessed with permission at https://pecantoolbox.nmsu.edu/.

Additional information

Communicated by E. Dirlewanger

Publisher’s note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Nolan Bentley and L. J. Grauke are co-first authors

Supplementary Information

Figure S1

Scatter and density plots visualizing the per marker maximum percentage of observations explained (%Exp) by an intercross phase (%Expinter) on the x-axis and the %Exp by a testcross phase while imputing homozygous minor allele calls as heterozygous (%Exptest) on the y-axis. The scatterplot shows all 15,429 SNPs meeting minimum MAF and missing call frequency filters colored by the frequency of heterozygous calls (Het-Freq). Loci with Het-Freq closer to 0.5 (the expected frequency for intercross and testcross markers) were plotted first to reduce overplotting. The density plots show the %Exp distributions for the 11,876 markers predicted based on the observed parental genotypes to be testcrosses (blue, median = 95.7%, median absolute deviation = 3.1%), intercrosses (red, median = 89.6%, median absolute deviation = 4.3%) and the not predicted (green). Red lines show the cutoff values used to accept or reject marker phasing; A) the minimum %Exptest at testcross loci (89.6%), B) the maximum %Exptest at intercross loci (84.7%), C) the minimum %Expinter at intercross loci and the maximum %Expinter at testcross loci (81.1%). Arrows D and E indicate distributions with low Het-Freq primarily explained by cross-types where the non-heterozygous parent contains two (D) or one (E) null alleles. This is indicated by their high homozygous minor-allele frequency and testcross-like segregation patterns. Arrow F points at unmappable AAxBB loci with high Het-Freq. Plotted with ggpubr (Kassambara 2017), colored with viridis (Garnier et al. 2018), and arranged with cowplot (Wilke 2016). (PNG 850 kb)

Figure S2

Line plots showing the relationship between marker physical position and linkage position across each chromosome. ‘Elliott’ (red), ‘VC1-68’ (blue), and consensus map (grey) positions plotted separately. Rug plots show the density of markers in each map in the corresponding color. The black rug plot indicates the location of all markers predicted to be mappable based on the observed parent genotypes. (PNG 1398 kb)

Figure S3

Pearson correlation between tree architecture phenotypes and associated measurements. Marker color based on position in the orchard (see row and column scatterplot). Red lines indicate line of best fit. Black line over histogram indicates density. p values of correlations (numbers) indicated as <0.05 (*), <0.01 (**), or < 0.001 (***). Correlations calculated and plotted using the psych (Revelle 2017) R package. (PNG 332 kb)

Figure S4

Pearson correlation between budbreak, pecan scab susceptibility, and associated measurements. Marker color based on position in the orchard (see row and column scatterplot). Red lines indicate line of best fit. Black line over histogram indicates density. p values of correlations (numbers) indicated as <0.05 (*), <0.01 (**), or < 0.001 (***). Correlations calculated and plotted using the psych (Revelle 2017) R package. (PNG 493 kb)

Figure S5

Scatterplot visualizing SNP marker positions comparing the results of the JoinMap v5 regression algorithm calculated linkage map (Table S5) to the R/QTL calculated consensus linkage map across all 16 chromosomes. All linkage groups calculated by JoinMap v5 consisted of markers from a single chromosome and parent. The JoinMap linkage maps were inverted when necessary to give the ‘Elliott’ informative markers a positive slope and the ‘VC1-68’ markers a negative slope to simplify visualization. Regressions calculated using lm function (Team RC 2016) and indicated by dark grey lines underneath the scatterplot. The r2 of this regression and the number of markers included (n) indicated per parental map. The overlying black lines show the reference sequence-based order of the markers and the color indicates the reference sequence position. (PNG 1058 kb)

Figure S6

Plots of LOD values associated with pecan scab incidence calculated in R/QTL (Broman et al. 2003) via the scanone Haley-Knott method with a step size of 0.1 cM. The line color and pattern indicate the origin of the plotted phenotype. Horizontal lines indicate the 98.75% cutoff for that trait determined by 10,000 permutations of the phenotypes. Rug plots show the relative LOD values (darker colors indicate greater LOD) in the ‘Elliott’ map (top rug plot), in the ‘VC1-68’ map (bottom rug plot), and at intercross loci in the consensus map (middle rug plot) plotted as a function of their position within the consensus map. (PNG 1196 kb)

Figure S7

Plots of LOD values associated with budbreak rating calculated in R/QTL (Broman et al. 2003) via the scanone Haley-Knott method with a step size of 0.1 cM. The line color and pattern indicate the origin of the plotted phenotype. Horizontal lines indicate the 98.75% cutoff for that trait determined by 10,000 permutations of the phenotypes. Rug plots show the relative LOD values (darker colors indicate greater LOD) in the ‘Elliott’ map (top rug plot), in the ‘VC1-68’ map (bottom rug plot), and at intercross loci in the consensus map (middle rug plot) plotted as a function of their position within the consensus map. (PNG 1166 kb)

Table S1

SSR primers, genotypes, and SNP marker associations (XLSX 56 kb)

Table S2

Marker statistics and phenotype associations (XLSX 8120 kb)

Table S3

Full calls and counts file for SNPs passing MAF and missing call frequency filters (XLSX 38210 kb)

Table S4

Results of tests comparing recombination frequencies between cross-direction sample groupings (XLSX 26632 kb)

Table S5

Re-analysis of marker phase and linkage position in JoinMap v5 performed across markers utilized in QTL mapping (XLSX 16 kb)

Table S6

Sample statistics and phenotypes (XLSX 47 kb)

Table S7

Summary of the scab incidence linear models and transformed marker analyses (XLSX 3368 kb)

Table S8

Summary of the budbreak linear models (XLSX 229 kb)

Table S9

Summary of the mqmscan results (XLSX 2405 kb)

Rights and permissions

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Bentley, N., Grauke, L.J., Ruhlman, E. et al. Linkage mapping and QTL analysis of pecan (Carya illinoinensis) full-siblings using genotyping-by-sequencing. Tree Genetics & Genomes 16, 83 (2020). https://doi.org/10.1007/s11295-020-01476-6

Download citation

  • Received:

  • Revised:

  • Accepted:

  • Published:

  • DOI: https://doi.org/10.1007/s11295-020-01476-6

Keywords

Navigation