Pedigree reconstruction and spatial analysis for genetic testing and selection in a Larix kaempferi (Lamb.) Carrière plantation

Abstract

Background Larix kaempferi is one of the major timber species in Northeast Asia. Demand for the reforestation of the species is rising in South Korea due to an increase in large timber production and utilization. However, progeny trials for the species have not been explored, making it challenging to foster advanced generations of tree improvement. In the present study, genetic testing and selection for diameter growth were conducted using pedigree reconstruction and phenotypic spatial distribution analysis in a plantation of L. kaempferi. The aim of the present study was to select the superior larch individuals using the pedigree reconstruction and phenotypic spatial distribution to substitute progeny trials. The plantation of seed orchard crops was established in 1990 and one-hundred and eighty-eight trees were selected as the study material. Genetic variation was investigated first to validate its adequacy as breeding material. Genetic testing was carried out using a model considering pedigree information and spatial autoregression of the phenotypes. Results The expected heterozygosity of the mother trees and offspring were 0.672 and 0.681 presenting the corresponding level of genetic variation between two groups. The pedigree reconstruction using maternity analysis assigned one to six progenies to ninety-two candidate mothers. The accuracy of genetic testing was exceedingly increased with the animal model considering AR1 ⊗ AR1 structure compared to the animal model only. The estimated genetic variance of the former was 9.086 whereas that of the latter was 4.9E-5 for DBH. The predicted breeding values of the offspring for DBH were ranged from -5.937 cm to 5.655 cm and the estimated heritability of diameter growth was 0.344. Conclusions The genetic testing approach based on pedigree reconstruction and phenotypic spatial distribution analysis was considered a useful analytical scheme that could replace or supplement progeny trials.

Principal component analysis revisited: fast multi-trait genetic evaluations with smooth convergence

“…This is a typical situation in breeding of species with long generation turnover, such as forest trees or some livestock animals. In these situations it could be more profitable to rather consider phenotypic profiles than individual traits with unclear connection to future breeding objective traits as diminishing age-age correlations reduces the response to selection [19, 41, 48]. As some traits are very expensive to measure, such as destructive sampling like meat quality traits in beef cattle [79], physiological traits in woody plants including fire-induced irreversible xylem damage and low temperature-induced tissue freezing [49], and wood (sawn timber) quality traits [26, 28], phenotype profiles could be measured and analysed with PCA to incorporate different types of traits jointly.…”