Impact of Negative Relationship between Yield and Protein Concentration in Pea Breeding for Plant-Based Protein Production

Authors: Daba, Sintayehu; McGee, Rebecca; Kiszonas, Alecia

Submitted to: ASA-CSSA-SSSA Annual Meeting Abstracts
Publication Type: Abstract Only
Publication Acceptance Date: 9/1/2022
Publication Date: 11/6/2022
Citation: Daba, S.D., Mcgee, R.J., Kiszonas, A. 2022. Impact of Negative Relationship between Yield and Protein Concentration in Pea Breeding for Plant-Based Protein Production. ASA-CSSA-SSSA Annual Meeting Abstracts. https://scisoc.confex.com/scisoc/2022am/meetingapp.cgi/Paper/141385.

Interpretive Summary: Separate breeding programs may be required to improve pea for different market classes. Pea breeding for plant-based protein isolation entails increasing grain yield and protein concentration. Negative associations between these two traits frequently make it difficult to improve them simultaneously. We used 36 yellow pea breeding lines, which were grown in two locations in Washington state (Fairfield and Pullman) in 2020 and 2021. The 2021 season was characterized by low rainfall and high temperatures, resulting in low grain yield of the lines. In terms of flour protein concentration (FLPC), the 2021 season was similar to or slightly better than the 2020 season. Protein isolate yield (PIYIELD) is a key trait for identifying pea cultivars for protein extraction, but it is time-consuming and laborious. As a result, PIYIELD evaluation is not feasible during the early stages of breeding, and it is critical to identify traits to use as indirect selection criteria. PIYIELD was shown to be favorably associated with FLPC (r = 0.67), suggesting that FLPC may be an essential criterion to breed peas for protein isolation. Grain yield (YIELD) had a moderate negative correlation coefficient with FLPC (r = -0.35). When the top 25% were selected based on BLUP of FLPC and YIELD, only G120 was identified by both criteria. The use of FLPC as a selection criterion resulted in a reduction of YIELD by 1.1% over the population mean. Selection based on YIELD resulted in gains for flour protein concentration as well. Most of the traits considered improved in a balanced manner using multi-trait selection indices (MTSI and MGIDI). Overall, the modest negative association between FLPC and YIELD could make it easier to find lines that mix the best of the two traits, especially when utilizing multi-trait selection indices.

Technical Abstract: Separate breeding programs may be required to improve pea for different market classes. Pea breeding for plant-based protein isolation entails increasing grain yield and protein concentration. Negative associations between these two traits frequently make it difficult to improve them simultaneously. We used 36 yellow pea breeding lines, which were grown in two locations in Washington state (Fairfield and Pullman) in 2020 and 2021. The 2021 season was characterized by low rainfall and high temperatures, resulting in low grain yield of the lines. In terms of flour protein concentration (FLPC), the 2021 season was similar to or slightly better than the 2020 season. Protein isolate yield (PIYIELD) is a key trait for identifying pea cultivars for protein extraction, but it is time-consuming and laborious. As a result, PIYIELD evaluation is not feasible during the early stages of breeding, and it is critical to identify traits to use as indirect selection criteria. PIYIELD was shown to be favorably associated with FLPC (r = 0.67), suggesting that FLPC may be an essential criterion to breed peas for protein isolation. Grain yield (YIELD) had a moderate negative correlation coefficient with FLPC (r = -0.35). When the top 25% were selected based on BLUP of FLPC and YIELD, only G120 was identified by both criteria. The use of FLPC as a selection criterion resulted in a reduction of YIELD by 1.1% over the population mean. Selection based on YIELD resulted in gains for flour protein concentration as well. Most of the traits considered improved in a balanced manner using multi-trait selection indices (MTSI and MGIDI). Overall, the modest negative association between FLPC and YIELD could make it easier to find lines that mix the best of the two traits, especially when utilizing multi-trait selection indices.