ARS | Publication request: Identification of informative marker panels for population assignment of rice germplasm

Submitted to: Rice Technical Working Group Meeting Proceedings
Publication Type: Proceedings
Publication Acceptance Date: February 22, 2010
Publication Date: February 22, 2010
Citation: Agrama, H.A., Yan, W., Mcclung, A.M. 2010. Identification of informative marker panels for population assignment of rice germplasm. Rice Technical Working Group Meeting Proceedings. CDROM Page 67.

Technical Abstract: Understanding the population sub-structure of rice germplasm accessions is useful for designing an effective strategy for gene transfer in a breeding program. Our objective was to develop a small marker panel that can effectively assign rice germplasm into one of five genetic sub-populations: indica, tropical japonica, temperate japonica, aus and aromatic. WHICHLOCI is a computer program that selects the best combination of loci for population assignment through empiric analysis of molecular marker data. Successive assignment trials using data from one locus at a time rank the loci based on their efficiency for correct population assignment, or discriminatory power. Subsequent trials with increasing numbers of loci are then performed to determine which combination contains the minimum number of loci required to reach a specific level of assignment success. The minimum number of loci with maximum assignment success for each genetic population is important for saving time and resources in a breeding program. The USDA rice germplasm core collection possessing 1,794 accessions was genotyped using 72 SSR markers. Each accession was inferred to one of the five sub-populations using the admixture analysis model-based clustering algorithms implemented in TESS v. 2.1. Using WHICHLOCI, each marker locus was ranked based on its discriminatory power for each sub-population, and a highly informative marker panel having a minimum number of loci required to reach 98% of assignment success was developed for each sub-population. Marker panels with 98% of assignment success included RM125, RM536 and RM555 for indica; RM11, RM489 and RM555 for tropical japonica; RM489, RM171 and RM145 for temperate japonica; RM178, RM551 and RM1339 for aus; and RM551, RM169 and RM408 for aromatic, respectively. Determination coefficient (r2) was 0.932 for indica, 0.910 for tropical japonica, 0.890 for temperate japonica, and 0.908 for aus. Thus, these initial results demonstrate that a set of 12 SSR markers was effective in assigning diverse germplasm to the five major sub-populations of rice.