Browsing by Subject "Genomic selection"

Now showing 1 - 10 of 10

Capturing wheat phenotypes at the genome level
(2022) Hussain, Babar; Akpınar, Bala A.; Alaux, Michael; Algharib, Ahmed M.; Sehgal, Deepmala; Ali, Zulfiqar; Aradottir, Gudbjorg I.; Batley, Jacqueline; Bellec, Arnaud; Bentley, Alison R.; Cagirici, Halise B.; Cattivelli, Luigi; Choulet, Fred; Cockram, James; Desiderio, Francesca; Devaux, Pierre; Dogramaci, Munevver; Dorado, Gabriel; Dreisigacker, Susanne; Edwards, David; El-Hassouni, Khaoula; Eversole, Kellye; Fahima, Tzion; Figueroa, Melania; Gálvez, Sergio; Gill, Kulvinder S.; Govta, Liubov; Gul, Alvina; Hensel, Goetz; Hernandez, Pilar; Crespo-Herrera, Leonardo Abdiel; Ibrahim, Amir; Kilian, Benjamin; Korzun, Viktor; Krugman, Tamar; Li, Yinghui; Liu, Shuyu; Mahmoud, Amer F.; Morgounov, Alexey; Muslu, Tugdem; Naseer, Faiza; Ordon, Frank; Paux, Etienne; Perovic, Dragan; Reddy, Gadi V. P.; Reif, Jochen Christoph; Reynolds, Matthew; Roychowdhury, Rajib; Rudd, Jackie; Sen, Taner Z.; Sukumaran, Sivakumar; Ozdemir, Bahar Sogutmaz; Tiwari, Vijay Kumar; Ullah, Naimat; Unver, Turgay; Yazar, Selami; Appels, Rudi; Budak, Hikmet
Recent technological advances in next-generation sequencing (NGS) technologies have dramatically reduced the cost of DNA sequencing, allowing species with large and complex genomes to be sequenced. Although bread wheat (Triticum aestivum L.) is one of the world’s most important food crops, efficient exploitation of molecular marker-assisted breeding approaches has lagged behind that achieved in other crop species, due to its large polyploid genome. However, an international public–private effort spanning 9 years reported over 65% draft genome of bread wheat in 2014, and finally, after more than a decade culminated in the release of a gold-standard, fully annotated reference wheat-genome assembly in 2018. Shortly thereafter, in 2020, the genome of assemblies of additional 15 global wheat accessions was released. As a result, wheat has now entered into the pan-genomic era, where basic resources can be efficiently exploited. Wheat genotyping with a few hundred markers has been replaced by genotyping arrays, capable of characterizing hundreds of wheat lines, using thousands of markers, providing fast, relatively inexpensive, and reliable data for exploitation in wheat breeding. These advances have opened up new opportunities for marker-assisted selection (MAS) and genomic selection (GS) in wheat. Herein, we review the advances and perspectives in wheat genetics and genomics, with a focus on key traits, including grain yield, yield-related traits, end-use quality, and resistance to biotic and abiotic stresses. We also focus on reported candidate genes cloned and linked to traits of interest. Furthermore, we report on the improvement in the aforementioned quantitative traits, through the use of (i) clustered regularly interspaced short-palindromic repeats/CRISPR-associated protein 9 (CRISPR/Cas9)-mediated gene-editing and (ii) positional cloning methods, and of genomic selection. Finally, we examine the utilization of genomics for the next-generation wheat breeding, providing a practical example of using in silico bioinformatics tools that are based on the wheat reference-genome sequence.
Evaluation of alternative statistical methods for genomic selection for quantitative traits in hybrid maize
(2012) Schulz-Streeck, Torben; Piepho, Hans-Peter
The efficacy of several contending approaches for Genomic selection (GS) were tested using different simulation and empirical maize breeding datasets. Here, GS is viewed as a general approach, incorporating all the different stages from the phenotypic analysis of the raw data to the marker-based prediction of the breeding values. The overall goal of this study was to develop and comparatively evaluate different approaches for accurately predicting genomic breeding values in GS. In particular, the specific objectives were to: (1) Develop different approaches for using information from analyses preceding the marker-based prediction of breeding values for GS. (2) Extend and/or suggest efficient implementations of statistical methods used at the marker-based prediction stage of GS, with a special focus on improving the predictive accuracy of GS in maize breeding. (3) Compare different approaches to reliably evaluate and compare methods for GS. An important step in the analyses preceding the marker-based prediction is the phenotypic analysis stage. One way of combining phenotypic analysis and marker-based prediction into a single stage analysis is presented. However, a stagewise analysis is typically computationally more efficient than a single stage analysis. Several different weighting schemes for minimizing information loss in stagewise analyses are therefore proposed and explored. It is demonstrated that orthogonalizing the adjusted means before submitting them to the next stage is the most efficient way within the set of weighting schemes considered. Furthermore, when using stagewise approaches, it may suffice to omit the marker information until the very last stage, if the marker-by-environment interaction has only a minor influence, as was found to be the case for the datasets considered in this thesis. It is also important to ensure that genotypic and phenotypic data for GS are of sufficiently high-quality. This can be achieved by using appropriate field trial designs and carrying out adequate quality controls to detect and eliminate observations deemed to be outlying based on various diagnostic tools. Moreover, it is shown that pre-selection of markers is less likely to be of high practical relevance to GS in most cases. Furthermore, the use of semivariograms to select models with the greatest strength of support in the data for GS is proposed and explored. It is shown that several different theoretical semivariogram models were all well supported by an example dataset and no single model was selected as being clearly the best. Several methods and extensions of GS methods have been proposed for marker-based prediction in GS. Their predictive accuracies were similar to that of the widely used ridge regression best linear unbiased prediction method (RR-BLUP). It is thus concluded that RR-BLUP, spatial methods, machine learning methods, such as componentwise boosting, and regularized regression methods, such as elastic net and ridge regression, have comparable performance and can therefore all be routinely used for GS for quantitative traits in maize breeding. Accounting for environment-specific or population-specific marker effects had only minor influence on predictive accuracy contrary to findings of several other studies. However, accuracy varied markedly among populations, with some populations showing surprisingly very low levels of accuracy. Combining different populations prior to marker-based prediction improved prediction accuracy compared to doing separate population-specific analyses. Moreover, polygenetic effects can be added to the RR-BLUP model to capture genetic variance not captured by the markers. However, doing so yielded minor improvements, especially for high marker densities. To relax the assumption of homogenous variance of markers, the RR-BLUP method was extended to accommodate heterogeneous marker variances but this had negligible influence on the predictive accuracy of GS for a simulated dataset. The widely used information-theoretic model selection criterion, namely the Akaike information criterion (AIC), ranked models in terms of their predictive accuracies similar to cross-validation in the majority of cases. But further tests would be required to definitively determine whether the computationally more demanding cross-validation may be substituted with the more efficient model selection criteria, such as AIC, without much loss of accuracy. Overall, a stagewise analysis, in which the markers are omitted until at the very last stage, is recommended for GS for the tested datasets. The particular method used for marker-based prediction from the set of those currently in use is of minor importance. Hence, the widely used and thoroughly tested RR-BLUP method would seem adequate for GS for most practical purposes, because it is easy to implement using widely available software packages for mixed models and it is computationally efficient.
Factors influencing the accuracy of genomic prediction in plant breeding
(2017) Schopp, Pascal; Melchinger, Albrecht E.
Genomic prediction (GP) is a novel statistical tool to estimate breeding values of selection candidates without the necessity to evaluate them phenotypically. The method calibrates a prediction model based on data of phenotyped individuals that were also genotyped with genome-wide molecular markers. The renunciation of an explicit identification of causal polymorphisms in the DNA sequence allows GP to explain significantly larger amounts of the genetic variance of complex traits than previous mapping-based approaches employed for marker-assisted selection. For these reasons, GP rapidly revolutionized dairy cattle breeding, where the method was originally developed and first implemented. By comparison, plant breeding is characterized by often intensively structured populations and more restricted resources routinely available for model calibration. This thesis addresses important issues related to these peculiarities to further promote an efficient integration of GP into plant breeding.
Genetic dissection of hybrid performance and heterosis for yield-related traits in maize
(2021) Li, Dongdong; Zhou, Zhiqiang; Lu, Xiaohuan; Jiang, Yong; Li, Guoliang; Li, Junhui; Wang, Haoying; Chen, Shaojiang; Li, Xinhai; Würschum, Tobias; Reif, Jochen C.; Xu, Shizhong; Li, Mingshun; Liu, Wenxin
Heterosis contributes a big proportion to hybrid performance in maize, especially for grain yield. It is attractive to explore the underlying genetic architecture of hybrid performance and heterosis. Considering its complexity, different from former mapping method, we developed a series of linear mixed models incorporating multiple polygenic covariance structures to quantify the contribution of each genetic component (additive, dominance, additive-by-additive, additive-by-dominance, and dominance-by-dominance) to hybrid performance and midparent heterosis variation and to identify significant additive and non-additive (dominance and epistatic) quantitative trait loci (QTL). Here, we developed a North Carolina II population by crossing 339 recombinant inbred lines with two elite lines (Chang7-2 and Mo17), resulting in two populations of hybrids signed as Chang7-2 × recombinant inbred lines and Mo17 × recombinant inbred lines, respectively. The results of a path analysis showed that kernel number per row and hundred grain weight contributed the most to the variation of grain yield. The heritability of midparent heterosis for 10 investigated traits ranged from 0.27 to 0.81. For the 10 traits, 21 main (additive and dominance) QTL for hybrid performance and 17 dominance QTL for midparent heterosis were identified in the pooled hybrid populations with two overlapping QTL. Several of the identified QTL showed pleiotropic effects. Significant epistatic QTL were also identified and were shown to play an important role in ear height variation. Genomic selection was used to assess the influence of QTL on prediction accuracy and to explore the strategy of heterosis utilization in maize breeding. Results showed that treating significant single nucleotide polymorphisms as fixed effects in the linear mixed model could improve the prediction accuracy under prediction schemes 2 and 3. In conclusion, the different analyses all substantiated the different genetic architecture of hybrid performance and midparent heterosis in maize. Dominance contributes the highest proportion to heterosis, especially for grain yield, however, epistasis contributes the highest proportion to hybrid performance of grain yield.
Genome-wide prediction of testcross performance and phenotypic stability for important agronomic and quality traits in elite hybrid rye (Secale cereale L.)
(2016) Wang, Yu; Miedaner, Thomas
Genomic selection offers a greater potential for improving complex, quantitative traits in winter rye than marker-assisted selection. Prediction accuracies for grain yield for unrelated test populations have, however, to be improved. Nevertheless, they are already favorable for selecting phenotypic stability of quality traits.
Genomic methods for rotational crossbreeding in local dairy cattle breeds
(2022) Stock, Joana; Bennewitz, Jörn
Local dairy breeds, such as German Angler, usually have small population sizes and thus a reduced genetic gain, compared to high-yielding breeds. Especially since genomic selection is widely used in the latter, the performance gap between local breeds and high-yielding breeds increased further, as it requires large reference populations in order to achieve accurate estimated breeding values. As a result, many farmers switched to high-yielding breeds. On the other hand, to increase the performance of local breeds the introgression of high-yielding breeds was a common strategy in the past, which resulted in high amounts of foreign genetic material in many of them. Much of the original genetic background got lost, however, they do not achieve the same performance level as high-yielding breeds. Local breeds are therefore faced with the risk of two types of extinction, i.e. a numerical extinction due to the small and decreasing numbers of breeding animals, and a genetic extinction due to massive introgression from high-yielding breeds. To promote local dairy breeds, the implementation of a genomic rotational crossbreeding scheme can be a promising strategy. Local breeds can benefit from a genomic rotational crossbreeding scheme with a high-yielding breed due to 1) an enlarged reference population including both the local breed and crossbred animals, and 2) the increased performance level of crossbred animals. On the other hand, crossbreeding is particularly known to improve functional traits by the exploitation of heterosis. Thus, it appears to be an appealling option for high-yielding breeds, as well, as they tend to struggle with fitness related problems. This thesis aimed to develop genomic methods for numerically small local dairy breeds in crossbreeding schemes in order to improve their genetic gain, genetic uniqueness, and their ability to compete with high-yielding breeds. In Chapter 2 a review study conducted a comparison of different genomic models which are suitable for crossbred data. Different additive models (such as the parental model, a model with breed-specific allele effects, and a single step model) and dominance models, which were either line-dependent, line-independent or included imprinting were discussed. It was concluded that the model choice needs to be made based on desired accuracies, computational possibilities, and data availability. In general, dominance models showed to result in higher accuracies compared to additive models. A breed of origin of alleles model approach was introduced in Chapter 3, which assumes different SNP effects for different origins of haplotypes. This model is suitable for the multi-breed genomic prediction of breeding values of numerically small breeds (i.e. German Angler) that have experienced introgression from high-yielding breeds in the past. The breed of origin of alleles model approach tended to be advantageous for Angler over multi-breed and within-breed genomic predictions with GBLUP. Chapter 4 contains a simulation study about the implementation of a rotational crossbreeding scheme including German Angler x German Holstein, while introducing genomic selection in Angler. Different sizes and structures of growing reference populations and selection goals of Angler were examined. The results showed that crossbred animals had a small overall superiority to both Holstein and Angler populations. In addition, a reference population containing both Angler and crossbred animals, in combination with a selection based on the purebred performance of Angler, gave the highest response to selection in the purebred Angler population and in the crossbred population. The difference between selection methods for Angler individuals could only be observed in the long term, as the purebred-crossbred correlations decreased. In Chapter 5 a simulation study on rotational crossbreeding was performed including different Optimum Contribution Selection methods, in order to realize genetic gain while regaining the original genetic background of Angler. Different constraints regarding mean kinships, native kinships, and migrant contributions from Holstein were applied to investigate their effects on Angler, crossbred, and Holstein populations. Constraining the amount of migrant contribution in Angler increased their genetic uniqueness. However, it led to a notable reduction of genetic gain and thus a reduced superiority of the crossbred animals. The slowed rate of genetic gain and thus the large difference of the performance between the parental breeds could not be compensated by heterosis effects. In Chapter 6 the thesis ends with a general discussion about further genomic models for crossbreeding, and the practical relevance of crossbreeding in dairy cattle.
Gibberella ear rot resistance in European maize : genetic analysis by complementary mapping approaches and improvement with genomic selection
(2022) Han, Sen; Melchinger, Albrecht E.
During the last decades, implementation of molecular markers such as single nucleotide polymorphisms (SNPs) has transformed plant breeding practices from conventional phenotypic selection to marker-assisted selection (MAS) and genomic selection (GS) that are more precise, faster and less resource-consuming. In this dissertation, we investigated these three selection approaches for improving the polygenic trait Gibberella ear rot (GER) resistance in maize (Zea mays L.), which is an important fungal disease in Europe and North America leading to reduced grain yield and grain contaminated with mycotoxins such as deoxynivalenol (DON) and zearalenone (ZON). Three different sets of materials were evaluated in multiple environments and analyzed for different objectives. In the first study, five flint doubled-haploid (DH) families (with size 43 to 204) inter-connected at various levels through common parents, were generated in an incomplete half-diallel design with four parental lines developed by the University of Hohenheim. Significant genotypic variances and generally high heritabilities were observed for all three traits (i.e., GER, DON and days to silking (DS)) in all families, implying good prospects for resistance breeding and phenotypic selection against GER across different environments in European maize germplasm. Genetic correlations were extremely tight between DON and GER and moderately negative for DS with DON or GER, suggesting that indirect selection against GER would be efficient to reduce DON, but maturity should be considered in GER resistance breeding. Using a high-density consensus map with 2,472 marker loci, we compared classical bi-parental mapping of QTL (quantitative trait locus/loci) with multi-parental QTL mapping conducted with joint families and using four different biometric models. Multi-parental QTL mapping models identified all and even further QTL than the bi-parental QTL mapping model conducted within each family. Interestingly, QTL for DON and GER were mostly family-specific, yet multiple families had several common QTL for DS. Many QTL displayed large additive effects and most favorable alleles originated from the highly resistant parent. Interactions between detected QTL and genetic background (family) were rare and had comparatively small effects. Multi-parental QTL mapping models generally did not yield higher prediction accuracy than the bi-parental QTL mapping model for all traits. In the second study, two diversity panels consisting of 130 elite European dent and 114 flint lines, respectively, from the University of Hohenheim were evaluated and subject to a genome-wide association study within each pool. Similar to the first study, highly significant genotypic and genotype × environment interaction variances were observed for GER, DON and DS. Heritabilities were moderately high for GER and DON and high for DS in both pools. Estimated genomic correlations between pools were close to zero for DON and DS, and slightly higher for GER. The detected QTL for DON were all specific to each heterotic pool and none of them was in common with previously detected QTL. Furthermore, no QTL was detected for GER and DS in both pools. Genomic prediction (GP) across pools yielded low or even negative prediction accuracy for all traits. When the training set (TS) size was increased by combining lines from both heterotic pools, the combined-pool GP approaches had no higher prediction accuracy than the within-pool GP approach. Different from expectation, method BayesB did not outperform genomic best linear unbiased prediction (GBLUP). In the third study, we analyzed two backcross (BC) families derived from a resistant and a susceptible recurrent parent. Both BC populations differed substantially in their means for all traits, suggesting that the two recurrent parents have different QTL alleles for GER resistance. Relatively high correlations were observed between DON and ZON concentrations measured by immunoassays and GER visual severity scoring and NIRS (near-infrared spectroscopy) within each BC population. Thus, the mycotoxin content in grain can reliably be reduced by directional selection for GER severity and NIRS measurements that are less expensive and less laborious. In conclusion, GER resistance in European maize germplasm can be effectively improved through breeding with resistant donor lines. GER visual severity scoring and NIRS measurements were found to be reliable predictors for DON and ZON concentrations in grain. We observed that QTL for GER and DON are mostly specific to a few families or a limited number of materials, whereas QTL for DS are more commonly shared between families. The multi-parental QTL mapping approach is complementary to the classical bi-parental QTL mapping in that the latter has generally higher power to identify rare but large-effect QTL for traits such as GER and DON, whereas the former is superior in detecting common but small-effect QTL for traits such as DS. Composing the TS with materials more closely related to the prediction set and increasing the TS size generally resulted in higher prediction accuracy for MAS and GS, irrespective of the trait and statistical model.
Investigations on methodological and strategic aspects of genomic selection in dairy cattle using real and simulated data
(2018) Plieschke, Laura Isabel; Bennewitz, Jörn
In Chapter one a method was developed to separate the genomic relationship matrix into two independent covariance matrices. Here, the base group component describes the covariance that results from systematic differences in allele frequencies between groups at the pedigree base. The remaining segregation component describes the genomic relationship that is corrected for the differences between base populations. To investigate the proposed decomposition three different models were tested on six traits, where the covariance between animals was described either only by the segregation component or by a combination of the two components. An additional variant examining the effect of a fixed modeling of the group effects was included. In total, 7965 genotyped Fleckvieh and 4257 genotyped Brown Swiss and 143 genotyped Original Braunvieh bulls were available for this study. The proposed decomposition of the genomic relationship matrix helped to examine the relative importance of the effects of base groups and segregation component in a given population. It was possible to estimate significant differences between the means of base groups in most cases for both breeds and for the traits analyzed. Analysis of the matrix of base group contributions to the populations investigated revealed several general breed-specific aspects. Comparing the three models, it was concluded that the segregation component is not sufficient to describe the covariance completely. However, it also was found that the model applied has no strong impact on predictive power if the animals used for validation show no differences in their genetic composition with respect to the base groups and if the majority of them have complete pedigrees of sufficient depth. The subject of the chapter two was investigation to systematically increase the reliability of genomic breeding values by integrating cows into the reference population of genomic breeding value estimation. For this purpose a dataset was generated by simulation resembling the German-Austrian dual-purpose Fleckvieh population.. The concept investigated is based on genotyping a fixed number of daughters of each AI bull of the last or last two generation of the reference population and, together with their phenotypic performance, to integrate them into the reference population of the genomic evaluation. Different scenarios with different numbers of daughters per bull were compared. In the base scenario the reference population was made up of 4200 bulls. In the extended scenarios, more and more daughters were gradually integrated in the reference population. The reference population of the most extended scenario contained 4200 bulls and 420,000 cows. It was found that the inclusion of genotypes and phenotypes of female animals can increase the reliabilities genomic breeding values considerably. Changes in validation reliability of 6-54% for a trait with a heritability of 0.4 depending on scenario were found. As the number of daughters increased, the validation reliability increased as well. It should be noted that the composition of the daughter samples had a very great influence on whether the additional genotyped and phenotyped animals in the reference population can have a positive effect on the reliability of genomic breeding values. If pre-selected daughter samples were genotyped, the mean validation reliability decreased significantly compared to a correspondingly large unselected daughter sample. In addition, a higher bias was observable in these cases. Chapter three expands the investigations of chapter two by a low-heritability trait, as well as the aspect of so called new traits. The results found in chapter two were confirmed in chapter three for a low-heritability trait. Changes in validation reliability of 5-21% for a heritability of 0.05 depending on scenario were found. The negative effects of pre-selected daughter samples were even more pronounced in chapter three. In the case of an ‘old’ trait, the number of phenotypes is expected to be (nearly) unlimited, since a recording system is well established. In the case of a new trait recording of phenotypes just started, therefore the number of phenotypes is limited. Two different genotyping strategies were compared for new traits. On the one hand, the sires of the phenotyped cows were genotyped and on the other hand the phenotyped cows were genotyped themselves. It was found in all compared scenarios that it is more sensible to genotype cows themselves instead of the genotyping their sires. However, if usual strategy of phenotyping female animals and genotyping of males is applied, it is at least important to ensure that many daughters are phenotyped in a balanced system. If different numbers of daughters per bull are phenotyped and unbalancedness becomes severe, the average validation reliability decreased significantly.
Optimum strategies to implement genomic selection in hybrid breeding
(2022) Marulanda Martinez, Jose Joaquin; Melchinger, Albrecht E.
To satisfy the rising demand for more agricultural production, a boost in the annual expected selection gain (ΔGa) of traits such as grain yield and especially yield stability has to be rapidly achieved. Hybrid breeding has contributed to a notable increment in performance for numerous allogamous species and has been proposed as a way to match the increased demand for autogamous cereals such as rice, wheat, and barley. An additional tool to increase the rate of annual selection gain is genomic selection (GS), a method to assess the merit of an individual by simultaneously accounting for the effects associated with hundreds to thousands of DNA markers. Successful integration of GS and hybrid breeding should go beyond the study of GS prediction accuracy and focus on the design of breeding strategies, for which GS maximizes ΔGa and optimizes the allocation of resources. The main goal of this thesis was to examine strategies for optimum implementation of GS in hybrid breeding with emphasis on estimation set design to perform GS within biparental populations and on the optimization of hybrid breeding strategies through model calculations. One strategy, GSrapid, with moderate nursery selection, one stage of GS, and one stage of phenotypic selection, reached the greatest ΔGa for single trait selection regardless of the budget, costs, variance components, and accuracy of genomic prediction. GSrapid was also the most efficient strategy for the simultaneous improvement of two traits regardless of the correlation between traits, selection index chosen, and economic weights assigned to each trait. The success of this strategy relies principally on the reduction of breeding cycle length and marginally on the increase in selection intensity. Moving from traditional breeding strategies based on phenotypic selection to strategies using GS for single trait improvement in hybrid breeding could lead not only to increments in ΔGa but also to large savings in the budget. The implementation of nursery selection in breeding strategies boosted the importance of efficient systems for inbred generation accompanied by improvements in the methods of hybrid seed production for experimental tests. When it comes to multiple trait improvement, the choice between optimum and base selection indices had minor impact on the net merit. However, considerable differences for ΔGa of single traits were observed when applying optimum or base indices if the variance components of the traits to be improved differed. The role of the economic weights assigned to each trait was determinant and small variations in the weights led to a remarkable genetic loss in one of the traits. The optimum design of estimation sets to perform GS within biparental populations should be based on phenotypic data, rather than molecular marker data. This finding poses major challenges for GS-based strategies aiming to select the best new inbreds within second cycle breeding populations, as breeding cycle length might not be reduced. Then, the ES design to optimize GS within biparental populations would have a defined application on the exploitation of within-family variation by increasing selection intensity in biparental populations with the largest potential of producing high-performing inbreds. Based on the results of this thesis, future challenges for the optimum implementation of GS in hybrid breeding strategies include (i) reductions in breeding cycle length and increments in selection intensity by refinements of DH technology or implementation of speed breeding, (ii) improvements in the methods for hybrid seed production, facilitating the reallocation of resources to the production of more candidates tested during the breeding cycle, and (iii) precise estimation of economic weights, reflecting the importance of the traits for breeding programs and farmers, and maximizing long term ΔGa for the most relevant traits.
Prospects of genomic selection for disease resistances in winter wheat (Triticum aestivum L.)
(2019) Grote, Cathérine Pauline; Miedaner, Thomas
Die Ziele dieser Arbeit waren (i) die erstmalige Evaluierung des Effekts des Zwerggens Rht24 auf FHB- und STB-Resistenzen, Wuchshöhe und Ährenschieben im Vergleich zum weit genutzten Locus Rht-D1, (ii) die Untersuchung des Potenzials der nichtadaptierten QTL Fhb1 und Fhb5 für die Entwicklung von Kurzstrohweizen, (iii) die Analyse der Vorhersagegenauigkeit von GS innerhalb und zwischen Familien durch die Anwendung der beiden Modelle RR-BLUP (ridge-regression best linear unbiased prediction) und wRR-BLUP (weighted RR-BLUP) und (iv) die Berechnung des Selektionsgewinns bzw. die Bestimmung der korrekt selektierten Top-10 %-Genotypen für FHB- und STB-Resistenzen durch GS. Die Ergebnisse dieser Studie zeigten, dass das gibberellinsäuresensitive Zwerggen Rht24 auf Chromosom 6 die Wuchshöhe um durchschnittlich 8,96 cm senkte, ohne dabei die FHB- und STB-Resistenzen oder den Zeitpunkt des Ährenschiebens ungünstig zu beeinflussen. Demgegenüber senkte das weitläufig verwendete Allel Rht-D1b die FHB-Resistenz um durchschnittlich 10,05 Prozentpunkte in einer Winterweizenpopulation bestehend aus acht biparentalen Familien, die für diese Resistenzloci segregierten. Diese Arbeit hat zusätzlich aufgezeigt, dass die Resistenzallele von Fhb1 und Fhb5 die FHB-Anfälligkeit um 6,54 bzw. 11,33 Prozentpunkte reduzierten und somit bereits allein das nicht-adaptierte Allel Fhb5b in der Lage ist, den negativen Effekt von Rht-D1b auf die FHB-Resistenz im untersuchten Material auszugleichen. Das verdeutlicht, dass die Wahl der Zwerg- und Resistenzgene in Zuchtprogrammen, in denen FHB-Resistenz ein Selektionsmerkmal ist, von entscheidender Bedeutung ist. In dieser Studie wurde des Weiteren das Potenzial der GS innerhalb und zwischen Familien untersucht. Die Vorhersagegenauigkeiten innerhalb einer Familie waren für alle Zielmerkmale höher als die zwischen Familien und unterschieden sich zwischen den einzelnen Familien und Vorhersagekonstellationen. Die stärkere Gewichtung von signifikanten Markern durch das wRR-BLUP-Modell führte zu einer Verbesserung der Vorhersagegenauigkeit im Vergleich zum weit genutzten RR-BLUP-Modell, wenn einzelne Gene, wie Rht-D1, oder Major-QTL, wie Fhb5, vorhanden waren. In dieser Studie wurden die genomisch geschätzten Zuchtwerte (GEBVs) von 2.500 ungeprüften Genotypen bestimmt, basierend auf einer partiell verwandten Trainingspopulation von 1.120 Genotypen. Die 10 % FHB- und STB-resistentesten Linien und eine zufällige Stichprobe wurden unter Berücksichtigung der Wuchshöhe genomisch selektiert und phänotypisch in einem vierortigen Feldversuch evaluiert. Für die FHB-Resistenz wurde ein Selektionserfolg von 10,62 Prozentpunkten relativ zur zufällig selektierten Populationsstichprobe ermittelt. Die GS erhöhte die STB-Resistenz allerdings nur um 2,14 Prozentpunkte. Auch die Selektion von neuen Kreuzungseltern auf der Basis von GS erscheint nicht ausreichend zuverlässig, da nur 19 % der Top-10 %-Individuen korrekt selektiert wurden. Zusammenfassend stellt die GS ein wertvolles Werkzeug dar, um den Zuchtfortschritt für die komplex vererbte FHB-Resistenz über kürzere Zyklen und größere Populationen zu unterstützen. In Kombination mit der Nutzung geeigneter Zwerggene und des nicht adaptierten QTL Fhb5 kann dadurch eine Steigerung der FHB-Resistenz im Winterweizen erzielt werden.