عنوان

Genetic Parameter Estimates and Prediction of Genetic Merit in Clonal Populations of Pinus taeda L. Using Expected and Realized Genomic Relationships

شماره

TL57712

زبان متن نوشتاري يا گفتاري و مانند آن

انگلیسی

عنوان اصلي

Genetic Parameter Estimates and Prediction of Genetic Merit in Clonal Populations of Pinus taeda L. Using Expected and Realized Genomic Relationships

نام عام مواد

[Thesis]

نام نخستين پديدآور

Shalizi, Mohammad Nasir

نام ساير پديدآوران

Holland, James B.

نام ناشر، پخش کننده و غيره

North Carolina State University

تاریخ نشرو بخش و غیره

2020

متن يادداشت

234 p.

جزئيات پايان نامه و نوع درجه آن

Ph.D.

کسي که مدرک را اعطا کرده

North Carolina State University

امتياز متن

2020

متن يادداشت

Quantitative genetics play a key role in the success of loblolly pine (Pinus taeda L.) tree improvement. Breeding and selection decisions are based on the outcome of genetic analyses of phenotype and more recently with DNA marker data. In this dissertation, we present statistical modelling of phenotype and DNA marker data to estimate genetic parameters and predict genetic merit of loblolly pine clonal varieties for growth, stem quality, and disease traits.  First, various variance-covariance structures were used to estimate genetic parameters and GxE interaction in a large multi-environment clonal population. Models with factor analytic additive genetic structures and spatially correlated residuals or polynomial fixed row and column effects efficiently captured heterogeneity in the data. The clone mean heritability estimate for stem volume was 0.41 for the simple compound symmetry additive genetic structure. The estimates were higher for more complex models, ranging from 0.56 to 0.61. Non-additive genetic variance for stem volume was a fraction of the additive genetic variance, suggesting that the trait is mainly controlled by large number of genes each with small effect. The results suggested that for forest trees genetic field tests with large number of genetic entries, breeders should consider incomplete row-column designs to model microsite heterogeneity in two directions.  Second, the correspondence between breeding values of 65 loblolly pine genotypes from clonal genetic tests and half-sib seedling progeny tests was studied. Additive genetic variance estimates from clonal tests were larger compared with the estimates from the half-sib progeny tests, regardless of the covariance structure used in the statistical models. Based on the independent analysis, the correlation between the breeding values of the same genotypes from two propagule types was moderate (0.59) for tree height and stem volume. The combined analysis resulted in a strong genetic correlation (>0.93) between the breeding values of two propagule types. Herein large discrepancy is mainly the outcome of different data analytical approaches. Conclusively, selecting genotypes for deployment based on clonal testing may not be optimal, but forest tree breeders can use the results from clonal tests to make some informed decisions.  Third, loblolly pine clonal varieties from five full-sib families were used to evaluate genomic selection for fusiform rust disease (caused by the fungus Cronartium quercuum f. sp. fusiforme) and stem forking defects. Ridge regression, Bayes B and Cπ models were implemented to study marker-trait associations and estimate predictive ability of SNP markers for selection. Three SNP loci had large effects on fusiform rust incidence, explaining 54% of variation in the trait. Cross-validation scenarios using a random set of 80% of the varieties for model building and 20% for validation had high (0.71- 0.76) prediction accuracies for both traits, while within-family and single family cross-validation scenarios produced low or weak prediction accuracies. When a family was split into training and validation set, the prediction accuracies were moderate to high (0.32 to 0.62) showing the importance of genetic connectedness. Genomic selection for fusiform rust disease incidence could be effective in loblolly pine breeding.  In the last chapter, genomic selection was investigated in a large loblolly pine clonal population using high density SNP markers. GBLUP, ridge regression, and Bayesian models were used to predict genomic estimated breeding values of clonal varieties within the same population or across test sites. Random sampling of individuals across population and within-families showed higher prediction accuracies compared with random sampling of families for model training and validation. The predictive ability of markers was moderate to high when data from tested varieties in seven sites were used to predict the same varieties a new environment. The predictive ability reduced when untested varieties were predicted across multiple environments. Genetic gain is expected to double with the application of genomic selection in loblolly pine breeding.

اصطلاح موضوعی

Botany

اصطلاح موضوعی

Developmental biology

اصطلاح موضوعی

Genetics

مستند نام اشخاص تاييد نشده

Shalizi, Mohammad Nasir

مستند نام اشخاص تاييد نشده

Holland, James B.

مستند نام تنالگان تاييد نشده

North Carolina State University

نام الکترونيکي

فرمت انتشار

p

نوع ماده

[Thesis]

کد کاربرگه

276903

سطح دسترسي

a

تكميل شده

Y