Genetic characterization and linkage disequilibrium estimation of a global maize collection using SNP markers

Abstract

A newly developed maize Illumina GoldenGate Assay with 1536 SNPs from 582 loci was used to genotype a highly diverse global maize collection of 632 inbred lines from temperate, tropical, and subtropical public breeding programs. A total of 1229 informative SNPs and 1749 haplotypes within 327 loci was used to estimate the genetic diversity, population structure, and familial relatedness. Population structure identified tropical and temperate subgroups, and complex familial relationships were identified within the global collection. Linkage disequilibrium (LD) was measured overall and within chromosomes, allelic frequency groups, subgroups related by geographic origin, and subgroups of different sample sizes. The LD decay distance differed among chromosomes and ranged between 1 to 10 kb. The LD distance increased with the increase of minor allelic frequency (MAF), and with smaller sample sizes, encouraging caution when using too few lines in a study. The LD decay distance was much higher in temperate than in tropical and subtropical lines, because tropical and subtropical lines are more diverse and contain more rare alleles than temperate lines. A core set of inbreds was defined based on haplotypes, and 60 lines capture 90% of the haplotype diversity of the entire panel. The defined core sets and the entire collection can be used widely for different research targets.

Figure 1. Allele frequency for total SNPs.

SNPs scored as biallelic markers are in the top panel, and scored as SNP haplotypes are in the bottom panel.

Figure 2. Estimated ln(probability of the data).

Ln(probability of the data) was calculated for K ranging from 1 to 10 (top panel) and delta K values for SNPs and SNP haplotypes for K from 2 to 9 (bottom panel).

Figure 3. Estimated population structure of the diverse inbred maize lines in the study.

Each of the 632 individuals is represented by a thin vertical line, which is partitioned into k colored segments that represent the individual estimated membership to the k clusters.

Figure 4. Distribution of pairwise relative kinship values.

Values equal to or greater than 0.40 were gouped as 0.40.

Figure 5. Linkage disequillibrium across the maize 10 chromosomes measured with 914 SNPs.

Only SNPs with a minor allele frequency greater than 0.05 are shown.

Figure 6. Mean LD estimates at different physical distances for three different minimum threshold cutoff levels for minimum allele frequency.

Mean LD estimates are pooled over all chromosomes, and three different minimum threshold cutoff levels for minimum allele frequency are shown.

Figure 7. Mean LD estimates for different physical distances for six different sample sizes.

Mean LD estimates are pooled over all chromosomes, and six different sample sizes using SNPs with minor allele frequency greater than 0.05 are shown.

Figure 8. Mean LD estimates at different physical distances for tropical and temperate subgroups.

Mean LD estimates are pooled over all chromosomes, and two subgroups (A = tropical, B = temperate) at sample size = 80 are shown, using SNPs with a minor allele frequency of 0.05.