Genetic variation and structure analysis of different type blueberries (Vaccinium spp.) based on SLAF-seq technology

Abstract: 【Objective】Blueberry belongs to the genus Vaccinium of family Ericaceae, and has recently received much attention as blueberries have high contents of anthocyanins, flavonols, procyanidins and other types of phenolic compounds, which can improve night vision, increase anti-cancer activity and reduce the risk of heart disease. The consumption and cultivation acreage of blueberries is thus continually increasing worldwide. By 2020, the total cultivated area of blueberries in the world was 2 056 700 hm2 and 66 400 hm2 in China, with a total production of 347 200 t. Blueberries are very rich in species, including the tetraploid highbush blueberries (Vaccinium corymbosum), lowbush blueberries (V. augustifolium) and hexaploid rabbiteye blueberries (V. ashei), and the diploid species V. uliginosum, V. vitisidaea, V. macrocarpon etc. However, their inter- or introgenetic background and phylogenetic relation-ship are still unclear, resulting in an unclear background of blueberry genetic diversity. The study could improve the understanding of the genetic basis of blueberries, by means of the genome-wide SNP markers to recognize the genetic variation and gene introgression between different blueberry populations. 【Methods】In this study, a total of 62 germplasm resources of 9 blueberry species were used as materials. The specific amplified fragment sequencing technology (SLAF-seq) was used to facilitate sequencing and accompanied by the reference genome of southern highbush blueberry draper. The sequencing reads were compared with the reference genome using BWA software (v 0.7.17), and data quality was controlled using GATK (v 4.1.7.0) and SAM tools, respectively; the intersection of SNP markers obtained by the two methods was used as the final reliable SNP marker dataset. Based on the final reliable SNP, the vcf tools software (v 0.1.16) was used to remove markers with a deletion rate over 20% or a MAF less than 0.05. The SNP hylo software (v 20140701) was used to filter SNPs with a depth of less than 3 and a deletion rate of less than 20% to obtain SNPs for further data statistics and analysis. First, the phylogenetic tree was drawn using the obtained SNP markers, and the validity and accuracy of the markers were verified by combining the phylogenetic tree and the traditional classification of the test materials. Then, the genetic structure of the test materials was analyzed using the Admixture software to understand the genetic background and individual genetic composition. To display the genetic background differences and genetic relationships of the test materials more intuitively, the SNP Relate (v 1.30.1) software package in R was used for Principal Component Analysis (PCA), and the ggplot2 (v 3.3.6) was used for visualization.【Results】We harvested 169.87 MB reads data, with an average Q30 value of 94.99% and an average GC content of 40.24%, and detected a total of 319590 SLAF tags in 62 blueberry materials. 172 513 polymorphic SLAF tags were verified by comparison with the reference genome. A total of 76 299 105 population SNP markers were detected using the GATK and SAM tools methods. A total of 4 133 595 population SNP markers were screened by integrity＞0.8 and minor allele frequency (MAF)＞0.05, 5.08% of the total SNP were evenly distributed among 12 chromosomes of blueberry. The phylogenetic tree was constructed based on the SNP markers and reflected the significant genetic differences among wild, northern highbush and southern highbush blueberry and the genetic relationships among cultivars, confirming the effectiveness and reliability of SNP markers in this study. The analysis of population structure based on SNP loci showed that there were significant genetic differences between wild types and cultivars, with the increase of K value, there was no subpopulation in wild types all the time, indicating that the tested wild types had independent genetic background and there was no gene exchange between the types, but there was unilateral gene flow from wild species to cultivars, suggesting that materials from wild species were used in the cultivation process of some cultivars, However, the cultivars had an obvious subgroup structure, suggesting that there was universal gene exchange between the subgroups. The classification of half- highbush blueberry Northland and southern highbush blueberry Reveille was closely clustered with the northern highbush type, The genetic background of the lowbush blueberry Blomidonand and half- highbush blueberry Northcountry and Blackpearl tended to be wild species at the genome-wide level.【Conclusion】The frequent use of backbone apparent varieties in breeding has increased the pattern of gene introgression between the bred cultivars, resulting in high genetic homogeneity in blueberries nowadays. The wild blueberry species are different in habitats and geographic distribution from the cultivars, and have an obviously different genetic background.