Characterization of Diversity in Traditional Northeastern Dry Bean Varieties and Potential for Genetic Improvement

Progress report for GNE19-208

Project Type: Graduate Student
Funds awarded in 2019: $14,932.00
Projected End Date: 03/31/2021
Grant Recipient: Cornell University
Region: Northeast
State: New York
Graduate Student:
Faculty Advisor:
Michael Mazourek
Cornell University
Expand All

Project Information

Project Objectives:

1) Assemble seedstocks of two traditional bean varieties (“Jacob’s Cattle” and “Arikara Yellow”) from 15 sources using regional seed companies and seed saver networks. These traditional varieties have been grown in the Northeast and throughout the United States for over 250 years. Our hypothesis is that decentralized cultivation of this landrace germplasm over many decades has led to regional adaptation and genetic divergence, creating a situation of low intra-population variation (i.e. within seed source) and high inter-population variation (i.e. between seed sources) within one culturally defined variety. This is supported by anecdotal evidence and variation in morphology.

2) Characterize genetic variance of collected landrace germplasm vs. commercial dry bean cultivars and analyze population structure. We will test the hypothesis that intra-population variance of the “crowdsourced” landraces will be greater than in commercial cultivars. Interestingly, a similar study of wheat landraces in France demonstrated just this result (6). These assembled populations of bean land races will form the basis of future population improvement and breeding work. Increased genetic variation of a population allows greater genetic gain through selection, and potentially greater stability of performance across different environments and abiotic and biotic stressors.

3) Evaluate agronomic traits of single variety-source plots in a replicated field trial at the Homer C. Thompson Vegetable Research Farm, using an augmented complete block design with four blocks. These data will complement our genotypic data by demonstrating what selectable variation may be present in the population of landrace germplasm for traits relevant to growers. A single replicate trial will also be planted in Maine at a cooperator farm of Fedco Seeds, but no phenotypic data will be collected at this site.

4) Conduct on-farm participatory selection and outreach with growers to select and improve diverse variety populations at collaborator site, Fedco Seeds. Fedco sells several traditional dry bean varieties, and has committed to hosting trial populations. This trial will take place at the farm of an experienced bean grower and long-time Fedco collaborator. These trials will serve as a “baby” site to the replicated trial at Cornell University, and will serve as the basis for a grower field day for dry bean growers and seed producers to observe the experimental population and select phenotypes of interest. This will serve to facilitate participatory methods of population improvement and begin the development of a regionally improved strain.


Edible dry beans were historically an important crop for the Northeast region of the United States, and as a legume species are beneficial for low-input, diverse crop rotation systems. Growers and consumers are increasingly interested in regionally produced staple crops, including traditional varieties valued for their culinary, visual and agronomic traits as well as their history. However, dry bean production in the Northeast has declined sharply over the past decades, and most variety improvement efforts focus on arid environments and production for the canning industry. At the same time, heirloom bean cultivars increasingly demanded by consumers lack important modern disease resistance and often are difficult to harvest mechanically. Expansion of cultivar options with high culinary and aesthetic value as well as favorable agronomic traits has the potential to increase productivity and profitability for growers and meet consumer demand for high quality edible beans.

The purpose of this project is to improve the viability and profitability of dry bean production in the Northeast by identifying superior genetic resources within two important traditional bean varieties, “Jacob’s Cattle” and “Calypso” (see Appendix: Figure 1), using genetic diversity information and participatory selection in order to improve current availability and provide material for future crop improvement.  In order to expand regional production of dry beans as a profitable crop, it is imperative to ensure that growers have access to varieties that meet their needs. This is attainable by selection within traditional germplasm, and by crossing to modern germplasm.

The secondary purpose is to convene growers of mixed experience levels to facilitate a passing of knowledge from experienced to new or prospective growers, especially regarding strategies for growing high quality bean seed in the Northeastern climate. 

Expansion of dry bean production in the Northeast as well as increased profitability of production, is important to maintaining healthy rural communities as well as diverse, healthy agro-ecological systems. It is also important to regional food security by providing regionally available sources of plant-based protein.


Click linked name(s) to expand
  • Heron Breen


Materials and methods:


    1. Two traditional bean varieties have been selected based on two qualifications: 1) relevance and potential for superior performance and adaptation in the Northeast, and 2) hypothesized intra-varietal diversity based on extent and location of cultivation. One recently released kidney bean cultivar, California Early Light Red Kidney, was selected to use as a control for the study, as they are the same race (race Nueva Granada, from the Andean center of origin) as the two traditional varieties selected. 
    2. We identified 15 unique sources for each landrace and obtain seed samples of at least 10 seeds from each (up to 16), with an effort to trace the origin of different sources and sample from distinct climatic environments. Methodology from Thomas et al. (2012) will be used to collect historical and geographical data on seed sources. Seed sources will be accessed from Fedco Seeds’ grower networks and seed repository in the Northeast, regional seed companies and the Seed Savers Exchange Network, which maintains a directory of thousands of varieties listed and shared by individual members. Fedco Seeds in Clinton, Maine is a collaborator on this project, and assisted with seed sourcing from its own collection, as well as through its grower networks. Seed was also sourced from the USDA-GRIN germplasm network. 
    3. (The fall season has been spent collecting germplasm, and preparations are being made to initiate the next steps that follow:)
    4. 5 seeds randomly selected from each source will be planted in the greenhouse for genotyping. Commercial cultivar seed was acquired from 5 commercial sources, as there are generally much fewer available commercial sources for modern cultivars. We hypothesize that intra-varietal diversity in the modern cultivars will be very low due to the practice of pure line and single plant selection for modern bean cultivars and the recent origins of the cultivar. Therefore, a smaller sample size will still adequately represent diversity present within the modern cultivars.  Due to sequencing costs, this unbalanced design is merited due to the goal of the project to thoroughly characterize genetic diversity present in the traditional varieties of interest, not the modern cultivars.
    5. Young plant tissue will be sampled at 14 days after planting, and DNA extraction will be done using established methods in the principle investigator’s laboratory facilities.
    6. Genotyping by Sequencing (GBS) methodology will be used to discover single-nucleotide polymorphisms (SNPs), an important measure of genetic diversity. This method is a widely used marker system for whole genome diversity studies and genetic mapping, and has been used successfully in both commercial and wild common bean(Corrado et al. 2017; Cortés and Blair 2018). This method was chosen over available SNP genotyping chips for common bean due to the reduced sampling power and potential for ascertainment bias for rare alleles of the latter method(Corrado et al. 2017). The GBS sequencing library will be prepared at University of Wisconsin-Madison Genomics Facility, and single-end Illumina sequencing will be conducted at the Cornell Biotechnology Center on the NextSeq500. SNP calling will be done using TASSEL software (v.5) using the available common bean reference genome(Schmutz et al. 2014). Diversity evaluation will be done using R (v. 3.3.3).
    7. Population structure analysis will be done in R as well as TASSEL. A haplotype network will be used to visualize relationships between germplasm sources(Thomas et al. 2012). Genetic diversity indices will be computed along with average number of alleles per locus, allelic richness, gene diversity, observed heterozygosity and population divergence. Associated SNPs will be referenced to identify potential known genes using PhytoMine and BioMart tools in Phytozome v.12(Cortés and Blair 2018).
    8. Use genotypic and population structure data to identify two plants per source that will represent maximum genetic variation of the population. These two plants per source will be grown to maturity in the greenhouse to provide seed for the subsequent replicated progeny row trials and phenotypic data collection.
    9. Field trials will be conducted at the Homer C. Thompson Research Farm (Freeville, NY) as well as grown at our collaborator trial site, Fedco Seeds, in a mother-baby trial design(Lyon et al. 2019). This trial design allows collection of research data from replicated plots at a centralized research location, as well as “baby” on-farm trial sites where only one replication is grown. Three replicates at the research farm site will be planted in a randomized complete block design. 
    10. Eight morphological traits will be measured: days to flowering (DAP), number of nodes, plant height (cm), pods in the upper 2/3 of the plant (%), yield per plant (g), 100- seed weight (g), and days to maturity (DAP). Number of nodes and plant height will be measured during the flowering time point; all other measured traits were evaluated at or after physiological maturity. Maturity will be measured by assessing the entire progeny row, and for all other traits, plants will be sampled from the inner 0.91 meters (3 feet) of each plant row.
    11. A Field Day in early September of 2020 to be held at the Fedco Seeds trial site will focus on on-farm selection and production of high quality bean seed. Dry bean growers and seed company representatives will be targeted will make selections based on progeny rows, with the goal of initiating regional variety improvement and selection as well as providing growers with tools for future on-farm variety maintenance and improvement. Seed of the population will be available to all original seed source donors, as well as other interested seed companies or seed growers.
Participation Summary
Any opinions, findings, conclusions, or recommendations expressed in this publication are those of the author(s) and do not necessarily reflect the view of the U.S. Department of Agriculture or SARE.