Classification of diverse pear genetic resources using a combination of MIG-seq and whole-genome sequencing

Pears ( Pyrus spp.), which are distributed across temperate regions of Eurasia, include several economically important species. Pear species and cultivar classification is always controversial, because hybridization occurs easily, due to a low reproductive barrier. To reveal the genetic diversity and genetic relationships among pear genetic resources, we applied multiplexed inter-simple sequence repeat genotyping by sequencing (MIG-seq) to accessions registered in the National Agriculture and Food Research Organization (NARO) Genebank, including several introduced from abroad and the wild individuals collected in Japan. In addition, whole-genome sequencing data from a public database were combined with MIG-seq data to improve the reliability of species and group classifications. Out of the 1018 accessions genotyped, we identified 97 synonym groups consisting of 267 accessions. After eliminating all but one accession from each synonym group, we identified some mislabeled and misclassified accessions by means of population structure analysis. The genetic clusters estimated by our population structure analyses revealed that our materials and criteria for pear group and species classification were similar to those used for the public database. Our analyses clarified that (1) P. ussuriensis var. aromatica arose from hybridization of Japanese pear ( P . pyrifolia ) and P. ussuriensis var. hondoensis , (2) the genetic composition of Japanese pear originated mainly from Chinese pear, whereas ancestral compositions from P. ussuriensis var. hondoensis were very limited in Japanese pear, and (3) P. pashia in Nepal and P. dimorphophylla , included in Asian pea pear groups, contain unique ancestral compositions that are genetically distinct from those of other Asian pea pear groups.