Understanding the Natural Qualitative Trait Relationships between Pima and Upland Cotton in the National Cotton Germplasm Collection

Friday, January 10, 2020: 8:15 AM
Brazos (JW Marriott Austin Hotel)
Amanda M. Hulse-Kemp , USDA-ARS
Daniel Restrepo-Montoya , North Carolina State University
Lori Hinze , USDA-ARS
Jodi Scheffler , USDA-ARS
Janna Love , USDA-ARS
Richard Percy , USDA-ARS
Don Jones , Cotton Incorporated
Candace Haigler , North Carolina State University
James Frelichowski , USDA-ARS
Quantitative phenotypic traits such as yield and fiber quality are variable across environments and may also be controlled by epistatic and pleiotropic effects. Contrastingly, qualitative morphological traits are relatively stable across normal growing environments and commonly used to evaluate genetic diversity. In the past two decades, molecular markers in genome mapping have largely superseded phenotypic surveys, however, qualitatively inherited traits continue to be used as sources for A) descriptors in cataloging accessions of germplasm collections, B) obtaining plant variety protection status, and C) germplasm registration. The current research focuses on comparative analysis of morphological traits of Upland Cotton (Gossypium hirsutum) and Pima Cotton (Gossypium barbadense). The USDA-ARS Crop Germplasm Research Unit recently updated a previous disparate descriptor schemes with a standardized rating scale that encompasses the diversity observed across Gossypium species. A large portion of the National Cotton Germplasm Collection has been evaluated under this standardized scheme and was analyzed here. A distribution and clustering analysis of categorical data using the collected standardized scores of 37 traits/descriptors, such as leaf hairs, boll nectaries, and seed type, among others, was performed. The traits evaluated have been collected over the last 6 years by the USDA-ARS at College Station, Texas, USA; Tecoman, Colima, Mexico; and Liberia, Costa Rica. The results allowed us to investigate the distribution of phenotypic traits and sets of clusters establishing similarities and dissimilarities across the evaluated dataset.