1:00:00 PM-1:15:00 PM
Non-GMO, waxy potatoes derived from CRISPR/Cas9-mediated mutations in the gbssI gene
Stephany Toinga-Villafuerte, Texas A&M University

               Starch is a polymer made of two polysaccharides: amylose and amylopectin. The ratio of these components determines industrial applications. Currently, the main source of starch in the market is corn, followed by cassava, potato, and wheat. Lately, the price of starch from corn has increased due to shortages and re-direction towards feed; therefore, an urgent need exists to find alternatives to native starch and also modified starches, such as waxy starch (amylose-free), for industrial applications like food amendment, for alcohol production, and as a coating for foods and drugs. Agrobacterium-mediated transformation was used to introduce the CRISPR/Cas9 components into the variety Yukon Gold (TXYG79) to target the granule-bound starch synthase gene that encodes the GBSSI enzyme, which is essential for amylose biosynthesis. The events obtained were examined for amylose/amylopectin ratios using perchloric acid and enzymatic methods. In addition, viscosity profiles of the tuber starch from six different knockout events were obtained using a Rapid Visco Analyzer. We successfully obtained one event, T2-7, with mutations in all four gbssI alleles, resulting in the waxy phenotype (amylose-free, pure amylopectin). We compared morphological and developmental characteristics of the gene-edited event T2-7 versus TXYG79 (wild-type). No detrimental characteristics were observed in the gene-edited event. It had similar tuber numbers, tuber weight per plant, and specific gravity. However, T2-7 is considered a genetically modified organism (GMO). Eliminating the CRISPR/Cas9-nptII components is necessary to obtain non-GMO waxy potatoes. T2-7 was selfed, crossed with Yukon Gold TXYG79, and NDTX081648CB-13W (a self-compatible chipper). The presence/absence of the CRISPR components was evaluated in the progeny from crosses. In the case of the selfed progeny of the T2-7, 10% of seedlings showed the absence of the Cas9-nptII cassettes while retaining the modified starch phenotype. Thus, gene-edited, non-GMO, waxy potato clones for potential industrial applications have been successfully obtained.

1:15:00 PM-1:30:00 PM
Mutation breeding as a tool for potato (Solanum tuberosum) varietal development
Oluwafemi Omilani, Oregon State University

               Mutation breeding remains a useful tool for plant breeders in crop improvement. Its efficiency, however, limits its widespread use today. Methods termed New Breeding Technologies (NBTs) like CRISPR, RNAi, etc., have proved more efficient, but their deployment in potato varietal development has been stagnated by myriad reasons. A better understanding of their optimal use efficiency can improve their adoption for breeding. Gamma radiation and ethyl methane sulfonate (EMS) are two popular physical and chemical mutagens. Gamma radiation is mostly thought to lead to chromosome-level alterations while EMS leads to single base substitutions. In this study, we attempted to investigate the relationship between the mutagen dose and the type of mutation induced. We also investigated their preponderance across the genome. To achieve this, we induced mutations in three potato clones (‘Umatilla Russet’, ‘Castle Russet’ and AO02183-2) by exposing one-week old internodal cuttings in vivo to seven gamma radiation doses (5, 10, 20, 30, 40, 50 and 60 grays), and twelve EMS doses (0.25%, 0.50%, 0.75% & 1% at 30 minutes, 1 hour & 2 hours). Genotypic analyses and phenotypic screening of mutants followed. The LD50 varied for the three clones using both mutagenic agents. Using SNP genotyping, our results suggest that gamma radiation expectedly led to INDELs, but it also induced SNPs in the mutant genomes at both low and high radiation doses. EMS expectedly induced SNPs in mutant genomes, but also induced INDELs at higher exposure concentrations. The distribution of the SNPs appears to be random for both the gamma radiation and EMS induced mutants, but investigations are underway to corroborate this. This information can guide mutation breeding efforts, and the resulting mutants could also serve as a valuable genetic resource which can be used to investigate associations between mutant loci and phenotypes for varietal development.

1:30:00 PM-1:45:00 PM
Exploration of mutant potato germplasm for use in breeding programs- clones with contrasting starch profiles
BENOIT BIZIMUNGU, Agriculture and Agri-Food Canada

               Potato is a major crop contributing to global food security and nutrition. As an excellent source of starch, potato has also many food and non-food industrial applications. In our previous study involving mutagenesis performed in diploid potato germplasm using ethyl methane sulfonate (EMS), we have produced numerous genotypes displaying a wide phenotypic variation in physiological, agronomic, and end-use quality traits. Current work aims to characterize suitable material for use in diploid or tetraploid breeding. Phenotyping evaluation has revealed genotypes with special starch profiles with potential value-added nutritional benefits or industrial applications. Genotyping characterization based on single nucleotide polymorphisms (SNPs) has also indicated the occurrence of unique allelic diversity in some genotypes. We present some unique germplasm and discuss favourable diversity that may contribute to potato breeding.

1:45:00 PM-2:00:00 PM
Predicting Carotenoid Content and Classifying Specialty Potatoes Using Near-Infrared Spectroscopy
Gustavo Henrique de Almeida Teixeira, University of Idaho     

               The yellow specialty potato cultivars have higher levels of carotenoids and -carotene, the most common carotenoid in food and the most potent provitamin A carotenoid. However, carotenoid determination in food products is a time-consuming destructive method that requires the use of chemicals, and sophisticated equipment. On the other hand, near-infrared (NIR) spectroscopy is a fast non-destructive method suitable for on-time/in-line applications. Although NIR spectroscopy has been applied to predict carotenoids in specialty potatoes, there is no information about the use of this technique to predict and classify the newly released Brazilian specialty cultivar ‘IAC Rurik’. Thus, the objective of this study was to develop prediction and classification models for carotenoid quantification in this cultivar. Commercially mature ‘IAC Rurik’ potato tubers were harvested in 2022 (three batches) and in 2023 (one batch). NIR spectra were acquired on the periderm of 200 tubers, in three positions, using a FT-IR Spectrum 100N spectrometer. Spectra were pre-processed (standard normal variate, first and second Savitzky–Golay derivatives). Potatoes were peeled, and tissues were sampled, immediately frozen in liquid N2, and the carotenoid content was determined using a UV-vis spectrophotometer. Partial least square (PLS) regressions were carried out for carotenoid prediction and Linear Discriminant Analysis (LDA) was applied for the classification of the tuber according to the harvests. Carotenoid content prediction was better obtained using spectra without pre-processing (RMSEC = 0.035 µg/g-1, Rc2 = 0.88, RMSEP = 0.036 µg/g-1, and Rp2=0.89). Tubers grown in the summer of 2022 grouped with those grown in the spring of 2023 and the other two harvests (autumn 2022 and spring 2022) were separated from each other using the second derivative of Savitzky-Golay rendering the highest accuracy prediction (86.5%). Thus, NIR spectroscopy can be used to predict carotenoid content in ‘IAC Rurik’ with low RMSEC, RMSEP, and high R2 values.

2:00:00 PM-2:15:00 PM
Strategies to improve early generation selection in potato breeding programs
David DE KOEYER, Agriculture and Agri-Food Canada

               Genetic improvement of cultivated potato (Solanum tuberosum L.) has several challenges, including autotetrasomic inheritance and high levels of heterozygosity. High-throughput, affordable molecular marker-assisted selection allows identification of progeny with desired traits early in the breeding cycle. Genomic prediction can identify better parents and variety candidates much earlier than approaches based solely on phenotypic assessment. Despite these advances, little progress has been made in improving the effectiveness of selection in the early generations of potato breeding programs. Typically, tens of thousands of seedlings are grown in a greenhouse to produce tuber families that are planted the next year in the field. Agriculture and Agri-Food Canada’s potato breeding program traditionally grew approximately 50,000 seedlings each year in un-replicated single hills. This represents a large portion of the land used by the program, from which relatively few (1-5%) seedlings advance to the next generation. Since 2019, we have split the families into replicated plots and have been transitioning to growing multiple hills per seedling. More recently, separate plots planted with extra mini-tubers from some families were used to compare tuber uniformity within and between families. These changes provide the framework for comparing parents and families for selection percentage, studying the genetic factors underlying selection in the early generations, and increasing the effectiveness of selection within potato breeding programs. Progress towards these goals will be highlighted in this presentation.

2:15:00 PM-2:30:00 PM
Leveraging multispectral derived relationships for phenomic and genomic selection in potato
Muyideen Yusuf, University of Minnesota

               High throughput phenotyping (HTP) has become an invaluable tool for increasing efficiency and precision in plant breeding. Multispectral leaf canopy reflectance from an unmanned aerial vehicle (UAV) provides information on physio-chemical properties of plants over a wide range of wavelength spectra and therefore can be utilized for enhanced phenomic selection in a similar approach to genomic selection. This could be a useful selection tool for potato breeding especially when phenotyping early generation trials with many entries. We compared three methods for generating estimated breeding values for our breeding program clones: first, using spectra derived relationships (W); second, using the traditional approach of genomic derived relationships (G); and third using a combination of both (W+G). Multispectral bands were collected at six different phenological growth stages for three different market class of potato: chips, fresh market, and russets. We modeled genetic main effects for yield and quality traits at each growth stage and for all stages combined. In general, the model with only the W kernel performed lower or similarly to the G only model depending on trait, while W+G kernels outperformed both single kernel predictions, in terms of accuracy irrespective of trait. We also investigated importance of each spectra variables through dimensionality reduction with all growth stages combined. This enabled the identification and use of selected spectra variables for improving the prediction ability with or without genomic data. This work highlights two potential uses for spectral data in genomic prediction, first, as an alternative to genetic data for making predictions and second in combination with genetic data to increase precision for when making selections.   

2:30:00 PM-2:45:00 PM
QTL analysis for tuber size, shape, and specific gravity in a tetraploid russet mapping population
Jaebum Park, USDA-ARS

               Attributes such as potato tuber size, shape, and specific gravity play pivotal roles in the quality, consistency, and yield of processed potato products, like French fries, within the russet market class. Thus, pinpointing genetic regions influencing these traits would significantly benefit potato breeding initiatives, particularly in advancing the development of molecular markers for marker-assisted selection. To gain deeper insights into the genetic basis of these traits, Quantitative Trait Locus/Loci (QTL) analysis was performed on a tetraploid mapping population from a cross between Palisade Russet and the breeding clone ND028673B-2Russ. For the measurement of tuber shape, two commonly used methods were employed: the length-width ratio, which is an objective method providing continuous numerical information, and the human eyes with a 5-point scale, which is a relatively subjective and intuitive approach. Additionally, tuber depth was measured since flattened tubers can negatively impact the recovery of French fries following processing. Specific gravity and tuber weight were also measured to indirectly evaluate the dry matter content and size of the tested tubers. MAPpoly and QTLpoly were used for autotetraploid linkage and QTL mapping, respectively. A major QTL mainly reflecting tuber length was consistently observed on chromosome 10. On chromosomes 4 and 6, relatively minor QTL associated with tuber shape were detected. Chromosome 2 had a significant QTL affecting tuber flatness. QTL for tuber weight (=size) and specific gravity were observed on chromosomes 5 and 3, respectively. The difference in QTL identification resulting from the choice of tuber shape measurement methods will also be discussed.

2:45:00 PM-3:00:00 PM             
Predictive Analytics of Russet Potato Selections from Oregon Potato Breeding Program
Fabiana Ferracina, Oregon State University     

               We explore the application of machine learning algorithms to predict the suitability of Russet potato clones for advancement in breeding trials. Leveraging data from manually collected trials in the state of Oregon, we investigate the potential of a wide variety of state-of-the-art binary classification models. We conduct a comprehensive analysis of the dataset that includes preprocessing, feature engineering, and imputation to address missing values. We focus on several key metrics such as accuracy, F1-score, and Matthews correlation coefficient (MCC) for model evaluation. The top-performing models, namely the multi-layer perceptron (MLPC), histogram-based gradient boosting classifier (HGBC), and a support vector machine (SVC), demonstrate consistent and significant results. Variable selection further enhances model performance and identifies influential features in predicting trial outcomes. The findings emphasize the potential of machine learning in streamlining the selection process for potato varieties, offering benefits such as increased efficiency, substantial cost savings, and judicious resource utilization. Our study contributes insights into precision agriculture and showcases the relevance of advanced technologies for informed decision-making in breeding programs.