ARS | Publication request: Modeling of Milling Yield Components and Their Relationship to Grain Quality QTLs in Long Grain japonica Rice

Submitted to: Plant and Animal Genome Conference Proceedings
Publication Type: Proceedings
Publication Acceptance Date: November 1, 2006
Publication Date: January 12, 2007
Citation: Kepiro, J.L., Fjellstrom, R.G., Mcclung, A.M., Yeater, K.M. 2007. Modeling of Milling Yield Components and Their Relationship to Grain Quality QTLs in Long Grain japonica Rice. Plant and Animal Genome XV, January 13-17, 2007, San Diego, CA. Available: http://intl-pag.org/15/abstracts.

Technical Abstract: Milling yield, also called "head rice recovery", is defined as the percentage of head rice obtained from rough rice. Milling yield is a critically important trait to the commercialization of rice cultivars because it largely determines the economic value of the farmer's crop. To investigate the inheritance of milling yield, a japonica recombinant inbred population segregating for traits used as selection criteria by U.S. breeders was evaluated for grain quality, grain appearance, and agronomic traits. SSR markers were used to anchor 532 AFLP markers in construction of a linkage map. Analysis of twenty-four traits identified sixty-five QTLs in sixteen chromosomal regions affecting these traits. Brown rice (BR), milled rice (MR), and head rice (HR) recovery were analyzed with 35 sub-component traits using forward stepwise regression modeling (SAS). Sixteen quality traits and six additional phenotypic traits were found to significantly affect milling yield. Forty-nine QTLs were significantly associated with these twenty-two traits. Some of the most important sub-components of milling were chalk, pale green kernels, and amylose content. The percent of the total variance explained by each of the models for BR, MR, and HR was 41.8%, 66.9% and 56.0%, respectively. Using simple regression, the proportion of brokens in the brown rice before milling (PB), explained 59.2% of the variance in HR, indicating this is an excellent predictor of milling yield. The largest QTL for PB, explaining 12.0% of the variance, was RM190 (waxy locus). Grain chalk significantly affected PB, and interestingly, RM190 was also the largest QTL for chalk.