SNP Identification from RAD-Seq Data in Faba Bean (Vicia faba L.)

Abstract

Faba bean is an important edible legume crop in China. Due to its huge genome size and no available reference genome, SNP marker is very limited in faba bean. To identify genome-wide SNP markers, we obtained 35.47 Gb data from eight landraces by RAD-sequencing, with an average of 4.77 Gb data for each accession. A total of 245443516 reads were generated, the single accession has an average of 30680439.5 reads, and the average length of the reads reaches to 144 bp. The Q20 and Q30 values were over 97.89% and 93.83%, respectively. The GC content between the reads varied from 38.05% to 40.09%. Using a special Bayesian method under the situation of no reference genome, we identified 3722 SNPs among the eight landraces. Regarding the single accession, the detected SNPs varied from 3278 to 3578, and the homozygous SNPs number was larger than that of heterozygous SNPs for most of accessions. For the SNP types, T:A->C:G type has the largest proportion (38.8%), followed by C:G->T:A (28.0%) and the smallest is T:A->A:T (7.5%). 31 SNPs were selected to convert into KASP markers, and they showed a success rate of 66.7% through amplifying on 46 accessions. The SNPs in this study provide a strong genetic tool for germplasm identification, gene mapping and molecular breeding in faba bean.