Liu et al. 2020 — GWAS of heavy metal ionomics in USDA mini-core rice collection
Liu et al. performed genome-wide association studies (GWAS) of sixteen ionomic traits — including cadmium (Cd), arsenic (As), nickel (Ni), and manganese (Mn) concentrations in rice grain — across 191 accessions from the USDA Rice Mini-Core Collection, grown under both flooded and unflooded conditions at Mississippi State University. Using 3.2 million SNPs and four GWAS methods (GLM, MLM, MLMM, FarmCPU), the study identified 106 QTLs for flooded ionomics, 47 for unflooded ionomics, and 97 for agronomic traits, of which approximately 80% were novel. The study establishes the genetic architecture underlying varietal differences in grain metal accumulation, providing a foundation for marker-assisted breeding toward lower-Cd, lower-As rice cultivars. Key candidate genes for Cd accumulation in grain were identified in QTL regions, consistent with known transporters such as OsNramp5 and OsHMA.
Key numbers
- 191 rice accessions from USDA mini-core collection
- 3.2 million SNPs used for GWAS
- 16 ionomic traits analyzed including Cd, As, Ni, Mn in grain (flooded and unflooded conditions)
- 106 QTLs identified for flooded ionomics; 47 for unflooded ionomics
- ~201 (~80%) of 250 total QTLs were newly identified
- Flooded conditions (standard paddy) consistently showed higher grain As and Cd than unflooded
Methods (brief)
Field experiment at Mississippi State University; grain ionomic concentrations measured by ICP-OES or ICP-MS (instrument not explicitly specified in abstract); GWAS performed with GLM, MLM (univariate), MLMM, and FarmCPU (multivariate) methods; Bonferroni threshold p < 1.53 × 10⁻⁸.
Implications
Certification: Provides genetic markers for breeding lower-Cd rice. Cultivar selection is a key lever for reducing grain Cd and As; this study gives the marker basis for identifying low-accumulator accessions. Courses: Demonstrates that variety × hydrology interaction (flooded vs. unflooded) is a major driver of grain metal variance — supports the geographic and agronomic variance framing. App: Reinforces that rice grain metal content varies substantially by cultivar, not just by origin geography.