2024
Poster Presentations
Assessing Potato Heat Tolerance Using Leaf Membrane Integrity
Amaka Ifeduba, Shuyang Zhen, Jeewan Pandey, M. Isabel Vales. Texas A&M University, College Station, TX.
Rising global temperatures are having an increasingly detrimental impact on agricultural food production, and potato, typically a cool-season crop, is no exception. Heat stress during the potato growing season leads to significant losses in tuber yield and quality. Since high-temperature episodes are difficult to predict and control, planting heat-tolerant varieties is recommended as an assurance to prevent reductions in yield and quality. In order to develop new heat-tolerant varieties suitable for different geographic areas and market preferences, breeders need to understand the distinctive mechanisms employed by heat-tolerant varieties and devise effective screening techniques to swiftly distinguish between heat-tolerant and heat-sensitive varieties. We are exploring membrane integrity in connection to heat tolerance. As a starting point, we compared the membrane integrity of five potato genotypes (‘Vanguard Russet’, ‘Reveille Russet’, ‘Sierra Gold’, ‘Russet Burbank’, and ‘Atlantic’) across a temperature range of 30 - 70 °C in 5 °C increments, each for 30 minutes, while measuring the electrolyte leakage at each temperature setpoint. The results revealed significant differences in relative electrolyte leakage among the genotypes, with the more heat-tolerant genotypes (Vanguard Russet and Reveille Russet) exhibiting significantly lower electrolyte leakage than the heat-sensitive Russet Burbank and Atlantic. We extended this investigation to a diverse panel of 217 clones from the Texas A&M potato breeding program. The clones were genotyped using the Infinium 22K V3 Potato Array and phenotyped to evaluate electrolyte leakage at 50 °C. Genome-wide association studies were carried out to identify the genomic regions linked to membrane integrity using the GWASpoly package. Genomic estimated breeding values were computed using Stagewise. Our study aims to provide greater insight into the genetic foundation of leaf membrane integrity and connection with heat tolerance, ultimately developing a rapid heat-tolerance selection tool to speed up breeding efforts toward the development of heat-tolerant potato varieties.
Genetic Control of Primocane Fruiting in Tetraploid Blackberry
Alexander Silva, Lacy Nelson, Carmen Johns, Ellen Thompson, Michael Hardigan, John Clark, Margaret Worthington. University of Arkansas, Hortifrut, USDA-ARS Corvallis.
Blackberry (Rubus spp.) is a specialty crop of increasing economic significance due to rising consumption, expanded marketing, and advancements in cultivar development. Only in the United States alone, blackberry production for the fresh market represents over $664 million. Blackberries, characterized by a perennial root system and biennial canes, have traditionally been categorized into floricane-fruiting (FF) cultivars, producing fruits exclusively in second-year canes, and the most recently improved primocane-fruiting (PF) cultivars, which can also produce fruits in first-year canes; a characteristic discovered in the diploid blackberry genotype ‘Hillquist’ (R. argutus). PF cultivars offer the potential for dual cropping in a single year, extending the production season, and exhibiting adaptability to tropical regions. Despite the economic importance of PF cultivars, the genetic basis of this trait has not been well elucidated. It has been proposed that PF is controlled by one major recessive locus, but its location in the blackberry genome is unclear. Here we used a set of more than 350 tetraploid blackberry genotypes to identify genomic regions associated with PF through a genome-wide association study (GWAS). A region located at 33 Mbp on chromosome Ra03 was highly associated with primo/floricane-fruiting variation. Within this genomic region, several annotated genes related to flowering, including an AP2-like gene and a Gibberellin 20 oxidase coding gene, were identified. To discover functional genomic variants linked to PF and identify candidate genes responsible for primocane-fruiting, we have been doing allele mining and transcriptome analysis. For allele mining, a set of 5 PF and 11 FF genotypes were resequenced and aligned to the R. argutus reference genome. We identified 365 SNPs that discriminated between PF and FF genotypes within a 5 Mbp window around the GWAS peak. From this group, eleven SNPs were selected to design KASP markers and used to genotype a diverse panel of 650 tetraploid blackberry genotypes from the University of Arkansas System Division of Agriculture (UADA), Hortifrut Genetics, and USDA Horticultural Crops Research Unit breeding programs. Two markers differentiated between PF and FF genotypes with an error rate lower than 5%. For transcriptome analysis, the youngest leaves and shoot apical meristems (SAM) from one FF and two PF blackberry genotypes were collected before and after the transition to the reproductive stage for RNA sequencing. This revealed a suite of differentially expressed flowering genes distributed across the entire genome, enlightening the potential molecular mechanisms influencing the flowering pattern between floricane and primocane-fruiting genotypes.
2023
Training presentations
Genetic marker imputation in polyploids - presented by Jeff Endelman
Comparisons of hybrid breeding strategies in clonal polyploids - presented by Marlee Labroo
User presentations
Tools and Techniques for GWAS and Genomic Prediction in Blueberries with Breeding Insight
Lillian Hilsop1, Michael Hardigan1, Moira Sheehan2, Dongyan Zhao2, Claire Luby1. 1USDA-ARS, Horticultural Crops Production and Genetic Improvement Research Unit, Corvallis, OR. 2Breeding Insight at Cornell University, Ithaca, NY.
Breeding Insight is a USDA initiative supporting underserved breeding programs with technologies and biotechnical expertise. The USDA-ARS Blueberry Breeding program in Corvallis, OR and Breeding Insight are teaming up to conduct genetic analysis of traits in a diverse population of blueberries adapted to the Pacific Northwest. This talk will outline the tools used in the analysis and the preliminary results. Five thousand genetically unique individuals from 85 bi-parental crosses were planted in Corvallis, OR in 2016 and 2017. In 2021, they were screened using the genetic marker platform developed by BI through Diversity Arrays Technology (DArT; Canberra, Australia). Marker data was cleaned and filtered using a UpDog, python, and R pipeline. From 2020-2022, phenotypes were collected using FieldBook software for plant physiology and fruit quality traits, such as berry size, acidity, firmness, obrix, and ripening time. Phenotypic data was stored in the BreedBase database. Genome-wide association and genomic prediction will be conducted using mapPoly, GWASPoly, and Stagewise. This presentation will share preliminary results, and the expected outcomes of the project. This work will benefit blueberry growers in the Pacific Northwest region where the blueberry industry is relatively new and growing rapidly by accelerating cultivar development and supporting public and private breeders with molecular breeding tools.
Genome-wide association mapping of fruit quality-related traits in a diversity panel of blueberries
Lauren Redpath2, Rish Aryal1, Amanda Hulse-Kemp4, Jaimie Green3, April Nyberg3, Nahla Bassil3, Hamid Ashrafi1. 1North Carolina State University, Raleigh, NC. 2Pairwise Inc., Durham, NC. 3USDA-ARS, National Clonal Germplasm Repository, Corvallis, OR. 4USDA-ARS, Genomics and Bioinformatics Research Unit, Raleigh NC.
Commercial blueberry is a perennial polyploid species that suffers from long generation times and a lack of advanced molecular breeding techniques. Southern breeding programs have prioritized interspecific hybridization and novel trait introgression. This study presents a comprehensive evaluation of eighteen fruit and agronomic traits paired with target capture genotyping utilizing 59,302 custom probes to identify trait-associated markers in a genome-wide association study (GWAS). Evaluated phenotypic traits included fruit size, weight, color, firmness, puncture-ability, soluble solids content, and titratable acidity, as well as agronomic traits of days to budbreak, full bloom, initial fruit maturity, and full fruit maturity. A total of 33,701 high-quality, high-confidence SNPs were identified, of which 9,855 SNPs aligned to twelve developed scaffolds, among which 21 significant SNPs were identified to have a trait association. Candidate genes were encoded by these SNPs with functions in anthocyanin acylation, firmness, fruit density, acidity, and flower development. The SNPs identified in this panel can be useful in the further identification of other unquantified novel characteristics. These data can be implemented in marker-assisted selection and development of advanced molecular breeding programs in blueberries.
The hidecan package: visual integration of GWAS and differential expression results
Olivia Angelin-Bonnet1, Matthieu Vignes2, Patrick Biggs2, Samantha Baldwin3, Susan Thomson3. 1The New Zealand Institute for Plant and Food Research Limited, Palmerston North, New Zealand. 2Massey University, Palmerston North, New Zealand. 3The New Zealand Institute for Plant and Food Research Limited, Lincoln, New Zealand.
Genome-wide association studies (GWAS) are often conducted with the goal of ultimately detecting genes involved in the genetic control of a trait of interest. Thus, in order to complement the results of such a study, differential expression (DE) experiments can be used to identify genes whose expression varies between two conditions. Alternatively, if several traits are measured for the genotyped individuals, it can be interesting to assess whether there is any overlap between the genomic regions found associated with each of the traits. In both cases, there is currently a lack of clear visualisation to summarise and integrate the results of multiple GWAS and/or DE analyses. Manhattan plots are only suited for a single GWAS, and circos plots do not permit a good comparison of the results from several analyses. Here we present the hidecan R package, for visualising the results from multiple GWAS and/or differential expression (DE) experiments. Given markers or genes scores from one or more GWAS and DE analyses, the package displays the genomic position of the markers and genes whose score exceeds a threshold set by the user. The significance threshold can be set independently for GWAS and DE results. In addition, the user can provide a list of candidate genes (e.g. genes known to affect the trait investigated), which will be added to the visualisation. The resulting hidecan plot provides a clear visual summary of the different analyses, and can integrate prior knowledge about the trait(s) under study. It can be used to identify genomic regions associated with a trait of interest both through high-scoring markers and differentially expressed genes or highlight any overlap in the genetic control of several related traits. The hidecan package also offers a Shiny app through which the user can create a hidecan plot, which can then be exported as a PNG or PDF file.
Poster presentations
Jeewan Pandey1, Sanjeev Gautam1, Douglas Scheuring2, Jeffrey Koym1, M. Isabel Vales2. 1Texas A&M University, College Station, TX. 1Texas A&M AgriLife Research and Extension Center, Lubbock, TX.
Malnutrition is a major public health problem in many parts of the world. This issue draws attention to all important sources of minerals for the human diet, including potatoes. Potato accounts for 18% of the recommended dietary allowances of potassium, 6% of copper, phosphorus, and magnesium; and 2% of calcium and zinc. Increased public interest in the nutritional value of foods has prompted the evaluation of mineral contents in potatoes and the investigation of the genetics underlying the traits. The objective of this study was to identify genomic regions associated with mineral content and obtain genomic-estimated breeding values to guide the selection of parents and the advancement of clones through the breeding pipeline in potatoes. A diversity panel with 214 potato clones was phenotyped in three field environments in Texas for 12 minerals and genotyped using the Infinium Illumina 22K V3 Potato SNP Array. A genome-wide association study for mineral content was performed using GWASpoly. Two QTLs on chromosome 7 associated with zinc content and three QTLs on chromosome 5 associated with potassium and manganese content were identified. The loci identified will contribute to a better understanding of the genetic basis of mineral content in potatoes and guide the enrichment of potato tuber with macro and micronutrients. Genomic-estimated breeding values were obtained using StageWise. Clones with high genomic-estimated breeding values should be used as parents to improve mineral content in potato tubers.
Lushan Ghimire, Norma Flor, Luis Felipe Ferrao, Juliana Benevenuto, Philip Harmon, Patricio Munoz. University of Florida, Gainesville, FL.
Bacterial wilt in blueberries is caused by Ralstonia solanacearum. Annual Florida grower’s surveys for breeding trait priorities in southern highbush blueberries (SHB) has ranked bacterial wilt among the most important diseases in terms of recent outbreak. Given the current rate of expansion of the Florida blueberry industry, several wilt outbreaks are unavoidable, as blueberries are being planted in areas where endemic pathogen populations exists. Therefore, we aim to accelerate phenotyping of SHB genotypes for disease response and to identify genomic regions conferring resistance to bacterial wilt using genome-wide association study (GWAS). The ultimate goal is to apply marker-assisted selection for accelerating the development of resistant cultivars. For this, softwood stems (12 inches) were collected from each individual in a population of 326 advanced selections (genotypes) from the University of Florida blueberry breeding program. The stems were dip-inoculated with bacterial suspension at a concentration of 1*10^8 CFU/mL for 10 minutes. They were then placed over a tray containing water and incubated inside a plant growth room at 28°C, with a photoperiod of 12 hours of light and 12 hours of darkness for a duration of two weeks. Disease severity of each genotype was visually assessed through the extent of wilting symptoms. A second phenotyping round was conducted on fifteen of each of the most resistant, average, and susceptible genotypes previously screened and on additional 204 genotypes from a new set of advanced selections. In this case, stem cuttings were continuously incubated in the inoculum (1 x 10^6 CFU/mL). Days to wilting and disease severity of each genotype were assessed. All these individuals were genotyped using Capture-Seq for SNP detection. Breeding values (BV) for the disease severity were calculated for both rounds using mixed models and pedigree information. BV for the first and second rounds of phenotyping ranged from 18.94 to 90.09 and from 17.21 to 93.4 respectively. Narrow sense heritability for the wilt-resistance trait was 0.54 and 0.36 as per the first and second rounds of phenotyping respectively. Further analyses are underway to identify significant genomic regions associated with bacterial wilt resistance in SHB and to better understand the genetic architecture of this trait. Additionally, computer vision phenomics is being explored for its potential in phenotyping SHB affected by bacterial wilt. Results from this research (obtained from screening and high-throughput genotyping) can ultimately aid in strategic breeding decisions and speed up the process associated with the development of bacterial wilt-resistant blueberry cultivars through marker-assisted selection.
Training presentations
2022
New software features for GWAS in polyploids - presented by Jeffrey Endelman: This presentation will cover new features added to GWASpoly in 2021, including (1) partial R2 values, (2) mixed ploidy datasets, and (3) VCF files as input.
2021
Overview of GWASpoly - presented by Jeffrey Endelman
2022
User presentations
Mason Chizk1, Margaret Worthington1, Carmen Johns1, Carly Godwin1, John R. Clark1, Rishi Aryal2, and Hamid Ashrafi2. 1University of Arkansas, Fayetteville, AR. 2North Carolina State University, Raleigh, NC.
Until recently, large-scale association mapping studies in autopolyploid crop species have been impractical due to cost of sequencing with adequate read-depth, computing limitations, and software availability. In a collaborative effort that leverages targeted next-generation sequencing (NGS) technologies and newly available software packages to overcome these obstacles, we report the first genome-wide association study (GWAS) in autotetraploid blackberry (Rubus L. subgenus Rubus Watson), identifying numerous quantitative trait loci (QTL) for potential use in marker-assisted selection (MAS). Using NGS via the Capture-Seq platform, 35,054 well-distributed probes were designed from the ‘Hillquist’ (Rubus argutus Link.) reference genome. From 2019 to 2021, a panel of 307 commercially available fresh-market blackberry cultivars and University of Arkansas Division of Agriculture breeding selections were sequenced and phenotyped for an array of traits that encompass fruit morphology, postharvest texture, acidity, sweetness, seed traits, thorniness, and plant height. Using the designed probe set, 81,150 high-quality single nucleotide polymorphisms (SNPs) were identified with greater than 150x average read-depth. Through phenotypic analyses, high broad-sense heritabilities exceeding 0.75 were observed for multiple quantitative traits including primocane height, thorn density, fruit weight, fruit length, drupelet number, 100-seed weight, and seed width/length ratio. QTL were identified for all for each of these traits except fruit weight. A QTL for fruit firmness was identified on Ra04, less than 0.5 Mb away from two polygalacturonase (PG) homologs and two pectin methylesterase (PME) homologs. Homologs for both candidate genes have been widely implicated in texture variation for other crops. A shared QTL for titratable acidity (TA) and pH was identified on Ra05 near three malate synthases, two vacuolar malate transporters, one MYB transcription factor, and one phosphoenolpyruvate carboxylase PEPC. Other major QTL associated with primocane height and thornlessness were discovered on chromosome Ra04 at 25 and 33 Mb, respectively. These data will not only be used to design the first diagnostic markers for MAS in blackberry, but will serve as a foundational training dataset for genomic selection to achieve gains in heritable traits with few to no high-impact QTL, such as soluble solid content (SSC) and fruit weight.
Nweze Nwabueze Peter1, Enoch G. Achigan-Dako2, Celestine Azubuike Afiukwa1, and Happiness O. Oselebe1. 1Ebonyi State University, Nigeria. 2University of Abomey Calavi, Benin Republic.
Black sigatoka disease caused by Pseudocercospora fijiensis is the most destructive disease of Musa species. The aim of this study was to identify Musa accessions cultivated in Benin Republic that possess host resistance capacity to black sigatoka disease. A total of 72 Musa accessions were used in the study including 58 local accessions and 14 accessions from International Musa Transit Centre (ITC) in Belgium. The experiment was laid out using the Augmented Split Plot Design at Misserete, in Benin Republic. Agronomic data were collected 9 months after planting according to standard evaluation method for the disease including; symptom appearance (SA, in days), rate of leaf death (%LD), youngest leaf spotted (YLS), number of standing leaves (NSL), index of non-spotted leaf (INSL), area under disease progress curve (AUDPC) and disease severity index (DSI). The presence of genes for black sigatoka resistance (bs1) was evaluated using a gene specific SSR marker and one unspecified disease resistant gene (UDRG) in Musa. DNA sequencing was done using ABI sequencer. The results revealed high significant difference (p<0.0001) in sigatoka disease incidence and severity among the accessions with the rate of symptom appearance (SA) ranging from 10 to 36 days, NSL and YLS ranged from 3.0 to 14.0 and 2.0 to 6.0 respectively, while the values of INSL, DSI and AUDPC ranged from 12.5 to 60.0, 13.33 to 52.0 and 58.0 to 239.0, respectively. A dendogram based on disease sensitivity data showed 3 distinct clusters (A-C). cluster A contains 20 resistant accessions, cluster B is made up of 16 susceptible accessions, while cluster C includes 36 intermediate resistant accessions. Also, the dendogram of the amplified resistance gene sequences and the gel matrix also clustered the Musa accessions into 3 groups which showed a little variation with from their morphological pattern of resistance to the disease. Multiple sequence alignment result based on UDRG sequences showed missing nucleotide sequence in some susceptible accession. The marker-trait association analysis revealed a mutual occurrence between UDRG_300 and the morphological indicators of resistance to the disease (YSL, NSL and INSL). The study therefore identified black sigatoka resistant Musa accessions among Benin Republic cultivars and provided a genetic tool (UDRG_300) which can be exploited for improvement of the crop for sigatoka disease resistance.
Genome-Wide Association for Drought Tolerance in Potato - presented by João Nomura et al.
Joao Nomura1, Moctar Kante2, Elisa Salas2, Hannele Lindqvist-Kreuze2, Fernando Angelo Piotto1, and Jeffrey Endelman3. !Escola Superior de Agricultura Luiz de Queiroz, Sao Paulo, Brazil. 2International Potato Center, Lima, Peru. 3University of Wisconsin - Madison, Madison, WI.
Potato is one of the most important crops for humans, especially in developing countries. Extreme climate events are becoming more frequent, and drought is one of the abiotic stresses that can severely impact yield. To understand the genetic response for drought tolerance in potato, 655 clones were evaluated at two semi-arid, lowland sites in Peru. All clones were evaluated under full and deficit irrigation at both sites. Water reduction in the deficit treatment began when 5% of the plants started to flower, which is a morphological sign of tuber initiation. The interval between irrigation events in the full treatment was 2-3 days vs. 15 days in the deficit treatment. Targeted amplicon sequencing (DArTag) was used for genotyping, which generated 2244 SNPs after quality control. Genotype calls were made using the R/updog package, and quantitative genetic analyses used the R/StageWise package. The mean yield under full vs. deficit irrigation was 533 vs. 204 g plant-1, which represents a yield loss of 62%. Across the panel of 655 clones, the yield reduction ranged from 42 – 84%. No marker-trait association was detected for yield under full irrigation, but under deficit irrigation we observed a GWAS peak at the beginning of chromosome 5 that accounted for R2 = 21% of the breeding value. This region corresponds to the location of CDF1, a well-known gene affecting potato maturity, and which more recently has been implicated in water homeostasis. When yield under drought was plotted against the visual ratings of vine maturity, there was a tendency for earlier clones to yield more than late clones. The average yield under drought increased by 21.5 g plant-1 per allele dose at the marker linked to CDF1. Even though drought tolerance is likely affected by many genes, CDF1 appears to be an important target for genetic improvement of this trait.
Poster presentations
Exploring the Genetic Control of Potato Tuber Dormancy - presented by Ao Jiao
Ao Jiao1, Sanjeev Gautam1, Jeewan Pandey1, Douglas C. Scheuring2, Jeffrey Koym1, and M. Isabel Vales1. 1Department of Horticultural Sciences, Texas A&M University, College Station, TX. 2Texas A&M AgriLife Research and Extension Center, Lubbock, TX.
Potato tuber dormancy is defined as the period after harvest during which tubers do not sprout even under favorable conditions. The length of the dormancy period is measured from vine kill until tubers start sprouting and it is affected by genotype and environmental factors. Premature dormancy break is a major factor causing post-harvest tuber quality reduction. Common methods to prevent sprouting include cold storage and the use of sprout inhibitors. Cold storage causes cold-induced sweetening and results in higher energy costs, whereas sprout inhibitors raise health and environmental concerns. Developing potato varieties with long dormancy could contribute to reducing the use of cold storage and sprout inhibitors. In this study, we evaluated tuber dormancy variation and investigated the genetic background of tuber dormancy. Over 200 clones from the Texas A&M Potato Breeding Program were grown in Dalhart, TX in 2019 and were evaluated for dormancy at room conditions (18 °C, RH 60%, dark). The clones exhibited variation in dormancy ranging from 38 to 155 days, with the Russets having significantly longer dormancy (> 96 days) than other market groups (70 – 80 days). Two Texas A&M varieties, Reveille Russet and Vanguard Russet, were among the clones with the longest dormancy. A genome-wide association study was performed using GWASpoly with Infinium Illumina 22K V3 Potato Array to identify genomic regions associated with tuber dormancy. The main QTL identified was on chromosome 9, explaining 11% of the phenotypic variation. Follow-up evaluations will be conducted at additional locations under room temperature and cold storage.
Sanjeev Gautam and M. Isabel Vales. Department of Horticultural Sciences, Texas A&M University, College Station, TX. Texas A&M AgriLife Research and Extension Center, Lubbock, TX.
Heat stress reduces marketable tuber yield and quality of potatoes. Tuber defects can be external (heat sprouts, chained tubers, knobs) or internal (vascular discoloration, internal heat necrosis). Successful cultivation of potatoes under heat stress requires planting heat-tolerant varieties that can produce high yields of marketable tubers. Heat tolerance is a complex trait and breeding for it possesses several bottlenecks. To facilitate future marker-assisted selection for heat tolerance, a genome-wide association study (GWAS) aimed to identify genomic regions associated with heat tolerance was conducted. Phenotyping for a panel of 217 diverse potato genotypes was conducted near Springlake, Texas (heat stress location) for two years using a randomized complete block design with two replicates. The genotypes differed in their capacity of expressing the external as well as internal defects on tubers under heat stress. GWAS was conducted using GWASpoly with Infinium Illumina 22 K V3 Potato Array. Significantly associated SNPs with external defect traits were located on chromosomes 3, 4, 6, and 7 while those with internal defect traits were located on chromosomes 3 and 10. The identified genomic regions may be important to improve heat tolerance in potatoes. Fine mapping of identified regions and validation of the markers associated with these regions would be required to further understand the mechanism involved in heat tolerance.