Fig 1.
Scatter plot of 413 accessions with 11,047 SNPs according to the first two principal components (PCs) for the 44k rice genome dataset.
IND: indica rice; TEJ: temperate japonica rice; TRJ: tropical japonica rice; AUS: Aus rice; AROMATIC: aromatic rice; ADMIX: admixed group.
Fig 2.
Flowchart showing the Monte Carlo simulation.
GEBV: genomic estimated breeding value; GBLUP: genomic best linear unbiased predictor; RIL: recombinant inbred line.
Table 1.
Ranking and GEBV averages (in parentheses) of the best F10 RILs selected by 30 repetitions of the seven proposed strategies applied to six traits in Dataset I.
Table 2.
Ranking and GEBV averages (in parentheses) of the best F10 RILs selected by 30 repetitions of the seven proposed strategies applied to three traits in Dataset II.
Fig 3.
GEBV averages of the best individuals selected from 30 repetitions at consecutive generations for the six chosen traits in Dataset I.
GEBV-O: subset of the top 10 accessions with minimal or maximal GEBVs; GD-O-30, -50, -100: subsets of 10 accessions with maximal D-scores chosen from candidate sets composed of the top 30, 50, and 100 accessions, respectively; GEBV-GD-30, -50, -100: subsets of the top two accessions plus eight accessions chosen from the remainder of the candidate sets composed of the top 30, 50, and 100 accessions, respectively, with maximal D-scores; BRSW: brown rice seed width; FPP: florets per panicle; FTAA: flowering time at Arkansas.
Fig 4.
GEBV averages of the best individuals selected from 30 repetitions at consecutive generations for the six chosen traits in Dataset I.
GEBV-O: subset of the top 10 accessions with minimal or maximal GEBVs; GD-O-30, -50, -100: subsets of 10 accessions with maximal D-scores chosen from candidate sets composed of the top 30, 50, and 100 accessions, respectively; GEBV-GD-30, -50, -100: subsets of the top two accessions plus eight accessions chosen from the remainder of the candidate sets composed of the top 30, 50, and 100 accessions, respectively, with maximal D-scores; FTAF: flowering time at Faridpur; PH: plant height; PNPP: panicle number per plant.
Fig 5.
GEBV averages of the best individuals selected from 30 repetitions at consecutive generations for the three target traits in Dataset II.
GEBV-O: subset of the top 10 accessions with minimal or maximal GEBVs; GD-O-30, -50, -100: subsets of 10 accessions with maximal D-scores chosen from candidate sets composed of the top 30, 50, and 100 accessions, respectively; GEBV-GD-30, -50, -100: subsets of the top two accessions plus eight accessions chosen from the remainder of the candidate sets composed of the top 30, 50, and 100 accessions, respectively, with the maximal D-scores; YLD, yield; PH, plant height; FT, flowering time.
Table 3.
GEBV averages of the best F10 RILs selected by 30 repetitions, based on the original group comprising 45 F1 hybrids and the reduced group comprising 17 F1 hybrids, using four strategies.
Table 4.
Average genetic gains from 30 repetitions for Dataset I.
Table 5.
Average genetic gains derived from 30 repetitions for Dataset II.