Linkage Mapping and QTL Analysis

Training presentations





  • MAPpoly, QTLpoly, and VIEWpoly updates: Use, progress and case studies - presented by Marcelo Mollinari, Gabriel Gesteira and Cris Taniguti. In this workshop, we will show how to construct genetic maps and perform QTL analysis in multiparental polyploid populations using new versions of the R packages MAPpoly and QTLpoly. In the hands-on section, we will use a tetraploid dataset to show the software features and present results in three interconnected hexaploid sweetpotato populations. We will also introduce VIEWpoly, an R package for visualizing and exploring results from polyploid computational tools using an interactive graphical user interface.
  • Polyploid Linkage and QTL Mapping with polymapR and polyqtlR - presented by Peter Bourke et al. In this workshop, participants will be re-acquainted with Wageningen’s polymapR / polyqtlR pipeline for linkage map construction and QTL analysis in polyploid F1 populations. Demonstration of new or extra software features not previously covered will be highlighted. These include using probabilistic genotypes in place of discrete genotypes, automatic building of multi-QTL models via co-factor analysis, visualizing meiotic pairing behavior, and genotype curation (spotting and correcting genotyping errors).



User Presentations

Poster Presentations

Candidate gene search for rose rosette disease resistance in the Rosa wichurana Basye’s Thornless genome assembly.

Tessa Hochhaus, Jeekin Lau, Cristiane H. Taniguti, David H. Byrne, and Oscar Riera-Lizarazu. Texas A&M University, College Station, TX.

Rose rosette disease (RRD) is one of the most threatening diseases to roses (Rosa sp.). RRD is caused by the rose rosette emaravirus and spread by an eriophid mite, Phyllocoptes fructiphilus Keifer. The disease symptoms include elongated and excessive branching of shoots, leaf distortion, red or yellow leaves, proliferation of prickles, and increased susceptibility to other stresses or diseases. Past studies have found quantitative trait loci (QTL) for reduced susceptibility to RRD on linkage groups (LGs) 1, 3, 5, and 6 in diploid populations, and LGs 1, 5, 6, and 7 in tetraploid populations. Recently, we utilized a meta-analysis approach to better understand the relationship between QTL identified in diploid and tetraploid populations. Through this meta-analysis, we were able to narrow the QTL regions. A meta-QTL on LG 5, MetaRRD5.2, and one on LG 6, MetaRRD6.1 were of particular interest due to the large proportion of phenotypic variance explained and the extent of colocalization of QTL. Meta-QTL intervals guided our search for genes related to antiviral mechanisms, R genes, and general plant defenses within the reference Rosa chinensis genome assembly. Since Rosa wichurana Basye’s Thornless is in the parentage of diploid populations used in this study and a genome assembly is now available, we performed a candidate gene search in the MetaRRD5.2 and MetaRRD6.1 intervals. With this, we hope to better define the gene content related to RRD resistance in these meta-QTL. 


QTL mapping of fruit quality traits in tetraploid kiwiberry (Actinidia arguta)

Ran Wang1, Xiujuan Qi2, Jinbao Fang2, Peter Bourke1, Richard G.F. Visser1, Chris Maliepaard1. 1Wageningen University and Research, 2Zhengzhou Fruit Research Institute.

Fruit quality traits play an important role in consumer preferences and consumption of kiwiberry (Actinidia arguta). The genetic basis of fruit quality traits in this woody, perennial and dioecious fruit crop so far remains largely unknown. This study aimed to identify the underlying genetic basis of fruit quality traits in A. arguta, using a single nucleotide polymorphism (SNP) genetic linkage map previously developed in a tetraploid F1 population of ‘Ruby-3’ x ‘KuiLv-M’. The F1 population was phenotyped over three years (2020–2022) for fruit quality traits, including skin colour, flesh colour, weight, diameter, total soluble solids, longitudinal diameter and fruit shape index. A total of nine QTLs were detected for five traits in this study, explaining between 10 - 32% of the trait variation. For fruit colour, the support interval of a major QTL on LG9 contained an MYB transcription factor MYB110, which was previously demonstrated to control colour regulation in kiwifruit, thus suggesting that the transcription factor MYB110 is the candidate gene for fruit colour in kiwiberry. The linked marker for fruit colour was validated in an F1 population and 24 kiwiberry cultivars. In conclusion, the knowledge obtained through the QTL mapping is applicable to improve the efficiency and cost-effectiveness in kiwiberry breeding.


User presentations

Genetic mapping and QTL analysis of the sex locus in tetraploid kiwiberry

Ran Wang, Chris Maliepaard, Peter Bourke, Jinbao Fang, Danny Esselink, Paul Arens, Roeland Voorrips, Richard G.F. Visser. Wageningen University and Research, the Netherlands.

The Actinidia genus is a perennial, dioecious, fruiting vine including 54 species. The main commercial crops are kiwifruit (Actinidia chinensis var. chinensis and Actinidia chinensis var. deliciosa) and kiwiberry (Actinidia arguta). The ‘kiwiberry’ is commonly referred to as the small, smooth-skinned, and hairless fruit. There are multiple ploidy levels in kiwiberry but the main cultivated cultivars are polyploid. Despite the availability of many genomic resources in kiwiberry, SNP genotyping is still a challenge given these different levels of polyploidy. Recent advances in SNP array technologies have offered a high-throughput genotyping platform of genome-wide DNA polymorphisms. In this study, we developed a high-density SNP genotyping array to facilitate genetic studies and breeding applications in kiwiberry. SNP discovery was performed by genome-wide DNA sequencing of 40 Actinidia genotypes. The identified SNPs were stringently filtered for sequence quality, predicted conversion performance, and distribution over the available diploid kiwifruit genome. A total of 134,729 unique SNPs were put on the array. The array was used to genotype a tetraploid kiwiberry F1 population using fitPoly. We constructed an integrated linkage map covering 3,060.9 cM across 29 linkage groups using polymapR. A high degree of collinearity was observed between the kiwiberry genetic map and the kiwifruit reference genome. Further, we performed QTL analysis for the sex locus using polyqtlR, and a single major QTL was identified on Linkage Group 3 (LG3) in kiwiberry. Finally, our investigation of chromosome pairing behaviour in the population showed that the kiwiberry is mostly tetrasomic inheritance with partial preferential pairing.


Discovery of major QTL conferring resistance to root-knot nematodes (Meloidogyne enterolobii, M. incognita) in the sweetpotato (Ipomoea batatas) mapping population ‘Tanzania’ x 'Beauregard' and development of KASP-based assays for marker-assisted breeding

Simon Fraher1, Tanner Schwarz1, Bonny Oloka1, Chris Heim1, Mercy Kitavi1, Gabriel de Siqueira Gesteira1, Marcelo Mollinari1, John Hamilton3, Guilherme da Silva Pereira2, Zhao-Bang Zeng1, Gina Brown-Guedira1, Adrienne Gorny1, C. Robin Buell3, G. Craig Yencho11North Carolina State University, Raleigh, NC. 2University of Vicosa, Vicosa, Minas Gerais, BR. 3University of Georgia, Athens, GA. 

Sweetpotato, Ipomoea batatas (L.) Lam. (2n=6x=90), is among the world’s most important food crops and it is North Carolina’s most important vegetable crop. The recent emergence of the plant parasitic nematode Meloidogyne enterolobii poses a significant economic threat to NC’s sweetpotato industry and breeding resistance into new varieties has become a high priority. Previous studies have shown that ‘Tanzania’, a released African sweetpotato landrace, demonstrated strong resistance toward the quarantined M. enterolobii as well as the economically significant but not quarantined M. incognita. We screened the biparental sweetpotato mapping population, ‘Tanzania’ x ‘Beauregard’, for resistance to M. enterolobii by inoculating 246 progeny each with 10,000 eggs under greenhouse conditions. ‘Tanzania’, the female parent, was highly resistant, while ‘Beauregard’ was highly susceptible. Bioassays exhibited strong skewing toward resistance for three measures of resistance: reproductive factor, eggs per gram of root tissue, and root gall severity rating. By all measures, a 1:1 segregation suggested a major gene conferred M. enterolobii resistance. Using a random-effect multiple interval mapping model, we identified a single major QTL, qIbMe-4.1, on linkage group 4 (57.5 cm) that explained 70% of the variation for resistance to M. enterolobii. A similar QTL was previously discovered in linkage group 7 for resistance to M. incognita, which we here describe as qIbMi-7.1. We aligned whole genome sequence data for ‘Tanzania’ and ‘Beauregard’ with the diploid I. trifida (NCNSP306) reference assembly to identify single dose SNP markers within these QTL. From a pool of 2,444 SNPs, we identified 35 as PCR-compatible and nearest to QTL peaks. DNA from 298 sweetpotato genotypes of known M. enterolobii and/or M. incognita resistance, including 122 TB progeny and 176 other diverse germplasm, was genotyped using competitive allele-specific PCR (KASP). Of the 35 putative markers genotyped, 20 amplified and were polymorphic. Bioassay and genotype data were compared using chi-square analyses to reveal four significant markers that predicted nematode resistance exceeding the NCSU sweetpotato breeding program’s requisite 80% accuracy. We report three markers for M. enterolobii resistance on linkage group 4: 7038814 with R2=0.22 (p=1.9e-05), 7038930 with R2=0.34 (p=2.6e-08), and 7039636 with R2=0.3434 (p=4.8e-08), and one marker for M. incognita resistance on linkage group 7: 1699509 with R2=0.34 (p=4.6e-08). Efficiency of these markers ranged from 89-93% concordance with bioassay data. Validation in additional germplasm may determine these markers to be suitable for what we believe to be the first implementation of marker-assisted breeding in sweetpotato.


Poster presentations

Identification of Quantitative Trait Loci Controlling Shoot Length, Internode Length, and Number of Internodes in Tetraploid Roses

Haramrit Gill, Jeekin Lau, Luke Whiteley, David Byrne, Oscar Riera-Lizarazu. Texas A&M University, College Station, TX. 

Plant height is an important factor in rose bush architecture that affects its aesthetic value. Roses with longer stems, for example, are desirable in the cut flower industry; however, roses with shorter stems and overall compact bush are desirable for landscape and garden roses. Plant height in tetraploid roses is complex and is affected by both internode number and length, ultimately affecting the overall shoot length. To better understand the genetic basis of this trait in roses, we conducted a study to identify quantitative trait loci (QTL) associated with shoot length, internode length, and the number of internodes. For this, measurements of shoot and internode length, and the number of internodes were collected for two years in two tetraploid rose populations [‘Stormy Weather’ (SW) X ‘Brite Eyes’ (BE), and ‘My Girl’ (MG) X ‘Brite Eyes’ (BE)]. Our analysis suggests the presence of a QTL on chromosome 3 for internode length and QTL on chromosomes 3 and 5 for shoot length and the number of internodes. The location of the QTL on chromosome 3 encompasses two genes. These were GID1c, involved in regulating vegetative growth, and RoKSN, a floral repressor and a regulator of vegetative growth in roses. So, we conclude that these genes could be the potential candidate genes responsible for regulating the internode length and hence the overall shoot length of the rose plant.


Identification of QTLs related to fruit texture at harvest and postharvest in blueberry

Heeduk Oh2, Molla Mengist2, Marti Pottorff, Lara Giongo3, Penelope Perkins-Veazie1, Massimo Iorizzo21Plants for Human Health Institute, Kannapolis, NC. 2North Carolina State University, Raleigh, NC. 3Edmund Mach Research and Innovation Center, San Michele, NC. 

Blueberries (Vaccinium corymbosum L.) are produced and consumed worldwide for their enjoyable flavor and numerous human health benefits. However, blueberry fruits are highly perishable and soften rapidly, negatively affecting postharvest transport, storage, and consumer acceptance. In this study, fruit characteristics related to pre- and post-harvest texture and berry size were evaluated in a mapping population including 346 F1 genotypes (‘Reveille’ x ‘Arlen’) over two years (2021 and 2022) and used for quantitative trait loci (QTL) analysis. Fruits were stored between 24 hours (T0) and 6 weeks (T1) at 2 C and 14 mechanical texture parameters were collected with a TA.XTPlus Texture Analyzer. Extensive variation was observed for all the texture and size parameters, indicating segregation for these parameters in this population. Fruit weight and the texture parameter, ‘distance to maximum force’ (distance the skin stretches until puncture), were found to be those most closely associated with postharvest texture changes. In total, eight and seven QTLs were detected for pre-harvest and post-harvest parameters, respectively, with some QTLs detected at both time points. A QTL on chromosome 10 for ‘force linear distance’ (a texture parameter identifying the length of curve between the point of contact and 80% strain) was stable over two years for both T0 and T1. These results are establishing a framework for developing a marker-assisted selection strategy for improving pre- and post-harvest texture characteristics in blueberry breeding programs.


Meta-analysis of rose rosette disease resistance quantitative trait loci and a search for candidate genes.

Tessa Hochhaus, Jeekin Lau, Cristiane Taniguti, David Byrne, Oscar Riera-Lizarazu. Texas A&M University, College Station, TX.

Rose rosette disease (RRD) caused by Rose rosette emaravirus is one of the major diseases on roses (Rosa sp.) which threatens the $168 million rose industry. The virus is transmitted by eriophyid mites, and symptomatic roses are generally destroyed to reduce further spread. The disease symptoms include elongated shoots, excessive branching of shoots, leaf distortion, red or yellow leaves, excessive prickles, and increased susceptibility to other stresses or diseases. Little is known about the genetics of RRD resistance, but recent quantitative trait mapping studies have revealed quantitative trait loci (QTL) for reduced susceptibility to RRD in linkage group (LG) 5, 6, and 7 in tetraploid, and LG 1, 3, 5, and 6 in diploid populations. In this study, we seek to better understand the relationship between QTL on LG 5 and 6 identified in both diploid and tetraploid populations and to better localize QTL by performing a meta-analysis using data from these reports. For this, we compiled and aligned genetic maps and projected meta-QTL. This analysis revealed that the peaks and intervals for QTL in LG 5 and 6 using diploid and tetraploid populations co-localized, suggesting that these may be the same QTL. The analysis also led to a narrowing of QTL intervals, which will guide the search for candidate genes. Results from this study may be used in the design of strategies to manipulate and use these genetic factors in a plant breeding context.


User presentations

Genetic Maze for polygenic traits in heterozygous polyploid kiwifruit population - presented by Jibran Tahir et al.

Jibran Tahir and Cyril Brendolise. The New Zealand Institute of Plant and Food Research. 

            Polyploidy is a key driver of significant evolutionary changes in plant species. The genus Actinidia (kiwifruit) exhibits multiple ploidy levels, which contribute to novel fruit traits, high yields, and resistance to the pathogen. Breeding programs focus on tetraploid cultivars, because of their resilience to the Psa disease, as well as robust fruit quality traits and size. However, there remains a substantial knowledge gap concerning the chromosomal biology and complex gene-trait associations in polyploid kiwifruit, which also exhibits high heterozygosity due to dioecy, compared with diploid species. Here we present the first case study where we employed CaptureSeq for genotyping 235 F1 heterozygous individuals of a tetraploid kiwifruit mapping population using 10K baits which generated a total of 725,175 raw SNP variants. Using this dataset we first identified caveats in the population structure by studying kinship and recombination landscape. We then performed dosage counts, using an empirical Bayesian analysis (Updog, “normal model”), and selected 45,436 high-quality SNPs across 188 F1s for generating ultra-dense linkage maps for 29 chromosomes, using polymapR. To further assess the quality of our maps, we estimated the Identity-by-descent (IBD) probabilities. We also tested the nature of pairing among chromosomes and provided the first evidence from a mapping population that the inheritance in polyploid kiwifruit species is mostly polysomic with preferential pairing in a few LGs. Finally, we performed QTL mapping for two key polygenic traits for breeding (bacterial resistance and flowering) and a monogenic trait for breeding operation (sex) and calculated the additive effects from various QTLs for a trait. We also tested an HRM-based SNP marker workflow downstream of QTL mapping, to assess the possibility of marker-assisted selection for these traits. Our study provides a detailed view of multiple challenges and successes in solving the genetic maze of polygenic traits in a polyploid dioecious species.


Identification of QTLs Controlling Resistance to Anthracnose Disease in Water Yam (Dioscorea alata) - presented by Paterne Agre et al.

Paterne Agre, Kumar Lava, Robert Asiedu, Asrat Asfaw. International Institute of Tropical Agriculture, Ibadan, Nigeria.

            Anthracnose disease, caused by the fungus Colletotrichum gloeosporioides, is the primary cause of yield loss in water yam (Dioscorea alata), the widely cultivated species of yam. Resistance to yam anthracnose disease (YAD) is a prime target in breeding initiatives to develop cultivars with durable resistance for the sustainable management of the disease in water yam cultivation. This study aimed at tagging quantitative trait loci (QTL) for anthracnose disease resistance in a bi-parental mapping population of D. alata. Parent genotypes and their recombinant progenies were genotyped using the GBS platform and phenotyped in two crop cycles for two years. A high-density genetic linkage map was built with 3,184 polymorphic SNP markers well distributed across the genome covering 1460.94 cm total length. On average 163 SNP markers were mapped per chromosome with 0.58 genetic distance between SNPs. Four QTL regions related to yam anthracnose disease resistance were identified on 3 chromosomes. The proportion of phenotypic variance explained by these QTLs ranged from 29.54 to 39.40%. The QTL regions identified showed genes that code for known plant defense responses such as GDSL-like Lipase/Acylhydrolase, Protein kinase domain, and F-box protein. The results from the present study provide valuable insight into the genetic architecture of anthracnose resistance in water yam. The candidate markers identified herewith form a relevant resource to apply marker-assisted selection as an alternative to a conventional labor-intensive screening for anthracnose resistance in water yam.

Poster presentations

QTL analysis of the “Tanzania” x “Beauregard” sweetpotato mapping population for resistance to Meloidogyne enterolobii - presented by Simon Fraher et al.

Simon Fraher1, Tanner Schwarz2, Bonny Oloka3, Ken Pecota1, Chris Heim1, Adrienne Gorny2, G. Craig Yencho1. 1Department of Horticultural Science, North Carolina State University, Raleigh, NC. 2Department of Entomology and Plant Pathology, North Carolina State University, Raleigh, NC. 3Root Crops Program, National Agricultural Research Organization, Entebbe, Uganda.

Due to its highly heterozygous hexaploid nature, sweetpotato (I. batatas) lags behind other crops in terms of genomic tools. New tools and strategies have afforded opportunities to associate sequence data with phenotypic data through QTL analysis, especially the publication of a diploid reference genome for I. trifida in 2018, and open-source software packages like QTLpoly and MAPpoly. QTL analysis is expected to be instrumental in accelerating breeding in this crop, notably for nematode resistance and nutritional factors. We analyzed the progeny of the sweetpotato biparental mapping population ‘Tanzania’ x ‘Beauregard’ (TB) representing 250 genotypes (including 4 check lines) for resistance to the emergent plant-parasitic nematode, Meloidogyne enterolobii (M.e.). Parent ‘Tanzania’ is highly resistant, and parent ‘Beauregard’ is highly susceptible. Bioassays showed clear bimodal segregation for resistance, suggesting a simplex major allele conferring resistance. Using the R package QTLpoly and the I. trifida reference genome, we discovered a major QTL peak at base pair 7,039,636 (79.21cM) of linkage group 4 of I. batatas associated with resistance to M.e. This analysis suggests variability in M.e. resistance within the TB population can largely be ascribed to genetic differences amongst the progenies (h² =66.9%). The next steps will search for flanking markers associated with these genotypes and attempt to identify markers that can be screened in the seedling stage, reducing the need for costly and laborious bioassays with this quarantined pest.


Octoploid Strawberry Linkage Map from Reduce Representation Sequencing SNP Markers - presented by Jose Guillermo Chacon

Jose Guillermo Chacon1, Marcelo Mollinari1, Bode A. Olukolu2, Zhao-Bang Zeng1, Gina E. Fernandez3. 1Department of Horticultural Science, North Carolina State University, Raleigh, NC. 2Department of Entomology and Plant Physiology, University of Tennessee, Knoxville, TN. 3Bioinformatics Research Center, North Carolina State University, Raleigh, NC. 

                The cultivated strawberry (Fragaria ×ananassa L.) is an allopolyploid (2n= 8x = 28) with a complex genomic composition that hindered genetic and genomic studies, as the similarity between subgenomes, introgressions from a dominant genome, and other issues complicate accurate mapping and variant calling. The recent publication of a full chromosome length reference genome and improvements of deep sequencing for polyploids facilitated overcoming part of those difficulties, including the development of linkage maps. A biparental population was generated crossing the NCSU selections NCS 10-080 × NCS 10-147. The parents and 280 seedlings were sequenced using the reduced representation sequencing OmeSeq protocol resulting in 2.47 billion reads. The ngsComposer application was used for quality control, demultiplexing, and filtering, resulting in 1.84 billion reads. The read alignment resulted in a coverage of 92.32% of the four subgenomes in the allo-octoploid strawberry reference genome ‘Camarosa’ 1.0. The map construction was done using the MAPpoly R package. After a quality control screening, a total of 6133 markers and 212 offspring individuals were used to build a genetic linkage map comprised of 28 linkage groups with a total length of 3154 cM and 4022 SNP makers. The minimum linkage group size was 30.2 cM and 49 markers, and a maximum of 155.69 cM and 168 markers, with an average interval genetic distance of 1.32 cm. The high marker density and the correspondence between the number of assembled linkage groups and the number of expected chromosomes indicates that MAPpoly R is a robust analysis. This analysis provides an excellent framework map for forthcoming studies, including QTL analysis and understanding modes of inheritance in this complex polyploid species.


Quantitative Trait Loci Associated with Flower Color Transition Phenotype in Tetraploid Roses - presented by Haramrit Gill et al. 

Haramrit Gill, Jeekin Lau, Qiuyi Fu, Natalie Anderson, David H. Byrne, and Oscar Riera-Lizarazu. Texas A&M University, College Station, TX.

                Flower color is one of the most important breeding traits in ornamental roses. The combination of particular anthocyanins, their co-factors, and their concentrations leads to different pigmentation patterns. We observed an interesting characteristic that we call ‘flower color transition’ in two tetraploid rose populations a) ‘Stormy Weather’ (SW) X ‘Brite Eyes’ (BE), b) ‘My Girl’ (MG) X ‘Brite Eyes’ (BE). The roses that exhibit this phenotype have flowers that transition from a light-yellow color to a dark pink/red (accumulation of anthocyanins) as the flower ages leading to bushes peppered with flowers of multiple colors. To our knowledge, the genetic control of this phenotype has not been studied in roses previously. Here, we present studies to better understand the inheritance of this trait and to identify quantitative trait loci (QTL) in two tetraploid bi-parental populations segregating for the ‘flower color transition’ trait. Our analysis suggests the presence of QTL on chromosomes 3 and 4. The location of QTL identified for the flower color transition coincides with the location of some genes involved in the flavonoid biosynthetic pathway. Additional studies are underway to validate these results.