A power analysis for detecting aging of dry-stored soybean seeds: Germination versus RNA integrity assessments

Authors: Tetreault, Hannah; FLEMING, MARGARET - Michigan State University; Hill, Lisa; Dorr, Emma; Yeater, Kathleen; Richards, Christopher; Walters, Christina

Submitted to: Crop Science
Publication Type: Peer Reviewed Journal
Publication Acceptance Date: 7/7/2022
Publication Date: 7/22/2022
Citation: Tetreault, H.M., Fleming, M.B., Hill, L.M., Dorr, E.J., Yeater, K.M., Richards, C.M., Walters, C.T. 2022. A power analysis for detecting aging of dry-stored soybean seeds: Germination versus RNA integrity assessments. Crop Science. 63(3):1481-1493. https://doi.org/10.1002/csc2.20821.
DOI: https://doi.org/10.1002/csc2.20821

Interpretive Summary: Currently genebanks rely on germination tests to monitor seed health of diverse germplasm collections. Germination assays tell whether a seed is alive or not, but does not predict when a seed deteriorates and dies. Therefore, the early subtle changes of deterioration cannot be detected without using thousands of seed, depleting a valuable collection. The current work builds from recent and related work on the use of RNA integrity (RIN) to predict seed health using a 32-year collection of soybean seeds with harvest years ranging from 1989 to 2019. In the current work we calculate the statistical power of RNA integrity assays (RIN) for monitoring seed aging during storage. RIN analyses were able to detect changes in stored soybean seeds faster using fewer seeds compared to traditional germination assays. Monitoring RNA quality presents an exciting new ability to predict seed health and increase efficiency of commercial and conservation seed storage operations.

Technical Abstract: Plant genebanks acquire large, diverse germplasm collections that deteriorate with storage time. Major challenges include maintaining good seed health in collections and detecting inevitable change during storage. Genebanks routinely rely on germination tests to monitor seed health; however, subtle signs of deterioration during the early stages of storage can only be detected by testing thousands of seeds. Assessment of RNA integrity (RIN) is an informative method that indicates seed aging rate before viability loss. In this study, we compared the statistical power of germination and RIN assays and quantified how sample size affects the ability to detect reduced germination or RIN. We used a legacy collection of Williams ’82 soybean seeds, with cohorts harvested between 1989 and 2019, to study deterioration during dry storage at 5°C. RNA was extracted and characterized from 2016 through 2021, providing an extensive dataset to assess effect-size directly over a five-year storage interval and indirectly over a 32-year storage interval. Combining models for aging rate, measured as either lost viability or declining RIN, allowed comparisons of effect-size for the two parameters with respect to storage time. During the early stages of seed aging, RIN decline was detected with smaller sample sizes over a shorter time increment, compared to lost viability. This study demonstrates that using RIN as a first and routine step in monitoring seed health, accompanied by less frequent germination assays, could help in seed conservation. Key words: aging, RIN, RNA integrity, germination, seed storage, viability testing, statistical power, longevity