A set of 134 SSR markers with high polymorphism covering the 17 linkage groups of pear were selected to study the genetic variability and relationships of 99 P. pyrifolia cultivars. A total of 660 allelic variants were detected, with the number of observed alleles per locus varying from 3 to 9, with a mean of 4.93 alleles per locus. Among 81 SSR markers newly developed from the pear genome, the highest polymorphism was obtained by the motif (GT/CA), and the least polymorphism generated by (AC/TG). Clustering relationships of 99 P. pyrifolia cultivars generally reflected geographical classification, and further confirmed the close relationship between pears from Yangtze River Basin and Japan. The results also indicated that Sichuan province played an important role in pear resource distribution. Genetic structure analysis revealed the similar relationship as clustering, indicating the close relationships between Japanese and Chinese pear; and further proved more genetic exchange among cultivars from Japan and Zhejiang, and fewer genetic exchange among cultivars from Yunnan, Hubei and Sichuan. CoreS1, a core collection of P. pyrifolia newly constructed by the preferred sampling strategy, included 24 cultivars, retained 95.74% observed number of alleles (Na) of the 99 P. pyrifolia cultivars. Principal component analysis (PCA) further supported the representative genetic diversity of CoreS1 for P. pyrifolia cultivars from China, Japan, and Korea. The high quality and comprehensive evaluation of P. pyrifolia cultivars by the SSR markers covering the whole genome demonstrates their potential application in genetic diversity, genetic relationship, and core collection research on other germplasm resources of pear.
We introduce new quantitative characteristics of the population using an analogy to the system of multi-spin molecules: the disease fields, which may depend on interactions, and the susceptibility to disease as derivative of genetic vector’s (GV’s) frequency of cases with respect to these fields. The genetic vector’s approach (GVA) is applied to statistical analysis of the interaction of two SNP haplotype of HTR2A and shared epitope (SE) alleles in relation to development of rheumatoid arthritis (RA). The analysis is performed for two independent cohorts, EIRA and NARAC, and based on the evaluation of double- and triple genotype–genotype versus SE alleles correlations. The Gibbs-like parametrization of GV frequencies makes analysis transparent and easy interpretable. We find that the main contribution into association to RA comes from GVs containing double SE. GVA may resolve an opposite role in risk/protection from different pairs of genetic variations and reveal an association to RA whereas the univariate analysis does not show significant association.