Ascertainment bias from imputation methods evaluation in wheat.

BACKGROUND: Whole-genome genotyping techniques like Genotyping-by-sequencing (GBS) are being used for genetic studies such as Genome-Wide Association (GWAS) and Genomewide Selection (GS), where different strategies for imputation have been developed. Nevertheless, imputation error may lead to poor performance (i.e. smaller power or higher false positive rate) when complete data is not required as it is for GWAS, and each marker is taken at a time. The aim of this study was to compare the performance of GWAS analysis for Quantitative Trait Loci (QTL) of major and minor effect using different imputation methods when no reference panel is available in a wheat GBS panel. RESULTS: In this study, we compared the power and false positive rate of dissecting quantitative traits for imputed and not-imputed marker score matrices in: (1) a complete molecular marker barley panel array, and (2) a GBS wheat panel with missing data. We found that there is an ascertainment bias in imputation method comparisons. Simulating over a complete matrix and creating missing data at random proved that imputation methods have a poorer performance. Furthermore, we found that when QTL were simulated with imputed data, the imputation methods performed better than the not-imputed ones. On the other hand, when QTL were simulated with not-imputed data, the not-imputed method and one of the imputation methods performed better for dissecting quantitative traits. Moreover, larger differences between imputation methods were detected for QTL of major effect than QTL of minor effect. We also compared the different marker score matrices for GWAS analysis in a real wheat phenotype dataset, and we found minimal differences indicating that imputation did not improve the GWAS performance when a reference panel was not available. CONCLUSIONS: Poorer performance was found in GWAS analysis when an imputed marker score matrix was used, no reference panel is available, in a wheat GBS panel.

Increased genomic prediction accuracy in wheat breeding through spatial adjustment of field trial data.

In crop breeding, the interest of predicting the performance of candidate cultivars in the field has increased due to recent advances in molecular breeding technologies. However, the complexity of the wheat genome presents some challenges for applying new technologies in molecular marker identification with next-generation sequencing. We applied genotyping-by-sequencing, a recently developed method to identify single-nucleotide polymorphisms, in the genomes of 384 wheat (Triticum aestivum) genotypes that were field tested under three different water regimes in Mediterranean climatic conditions: rain-fed only, mild water stress, and fully irrigated. We identified 102,324 single-nucleotide polymorphisms in these genotypes, and the phenotypic data were used to train and test genomic selection models intended to predict yield, thousand-kernel weight, number of kernels per spike, and heading date. Phenotypic data showed marked spatial variation. Therefore, different models were tested to correct the trends observed in the field. A mixed-model using moving-means as a covariate was found to best fit the data. When we applied the genomic selection models, the accuracy of predicted traits increased with spatial adjustment. Multiple genomic selection models were tested, and a Gaussian kernel model was determined to give the highest accuracy. The best predictions between environments were obtained when data from different years were used to train the model. Our results confirm that genotyping-by-sequencing is an effective tool to obtain genome-wide information for crops with complex genomes, that these data are efficient for predicting traits, and that correction of spatial variation is a crucial ingredient to increase prediction accuracy in genomic selection models.

High-throughput phenotyping and genomic selection: the frontiers of crop breeding converge.

Genomic selection (GS) and high-throughput phenotyping have recently been captivating the interest of the crop breeding community from both the public and private sectors world-wide. Both approaches promise to revolutionize the prediction of complex traits, including growth, yield and adaptation to stress. Whereas high-throughput phenotyping may help to improve understanding of crop physiology, most powerful techniques for high-throughput field phenotyping are empirical rather than analytical and comparable to genomic selection. Despite the fact that the two methodological approaches represent the extremes of what is understood as the breeding process (phenotype versus genome), they both consider the targeted traits (e.g. grain yield, growth, phenology, plant adaptation to stress) as a black box instead of dissecting them as a set of secondary traits (i.e. physiological) putatively related to the target trait. Both GS and high-throughput phenotyping have in common their empirical approach enabling breeders to use genome profile or phenotype without understanding the underlying biology. This short review discusses the main aspects of both approaches and focuses on the case of genomic selection of maize flowering traits and near-infrared spectroscopy (NIRS) and plant spectral reflectance as high-throughput field phenotyping methods for complex traits such as crop growth and yield.

Mapping of barley homologs to genes that regulate low temperature tolerance in Arabidopsis.

We investigated the allelic nature and map locations of Hordeum vulgare (barley) homologs to three classes of Arabidopsis low temperature (LT) regulatory genes-CBFs, ICE1, and ZAT12-to determine if there were any candidates for winterhardiness-related quantitative trait loci (QTL). We phenotyped the Dicktoo x Morex (DxM) mapping population under controlled freezing conditions and in addition to the previously reported 5H-L Fr-H1 QTL, observed three additional LT tolerance QTLs on 1H-L, 4H-S, and 4H-L. We identified and assigned either linkage map or chromosome locations to 1 ICE1 homolog, 2 ZAT12 homologs, and 17 of 20 CBF homologs. Twelve of the CBF genes were located on 5H-L and the 11 with assigned linkage map positions formed 2 tandem clusters on 5H-L. A subset of these CBF genes was confirmed to be physically linked, validating the map position clustering. The tandem CBF clusters are not candidates for the DxM LT tolerance Fr-H1 QTL, as they are approximately 30 cM distal to the QTL peak. No LT tolerance QTL was detected in conjunction with the CBF gene clusters in Dicktoo x Morex. However, comparative mapping using common markers and BIN positions established the CBF clusters are coincident with reported Triticeae LT tolerance and COR gene accumulation QTLs and suggest one or more of the CBF genes may be candidates for Fr-H2 in some germplasm combinations. These results suggest members of the CBF gene family may function as components of winter-hardiness in the Triticeae and underscore both the importance of extending results from model systems to economically important crop species and in viewing QTL mapping results in the context of multiple germplasm combinations.

Structural, functional, and phylogenetic characterization of a large CBF gene family in barley.

CBFs are key regulators in the Arabidopsis cold signaling pathway. We used Hordeum vulgare (barley), an important crop and a diploid Triticeae model, to characterize the CBF family from a low temperature tolerant cereal. We report that barley contains a large CBF family consisting of at least 20 genes (HvCBFs) comprising three multigene phylogenetic groupings designated the HvCBF1-, HvCBF3-, and HvCBF4-subgroups. For the HvCBF1- and HvCBF3-subgroups, there are comparable levels of phylogenetic diversity among rice, a cold-sensitive cereal, and the cold-hardy Triticeae. For the HvCBF4-subgroup, while similar diversity levels are observed in the Triticeae, only a single ancestral rice member was identified. The barley CBFs share many functional characteristics with dicot CBFs, including a general primary domain structure, transcript accumulation in response to cold, specific binding to the CRT motif, and the capacity to induce cor gene expression when ectopically expressed in Arabidopsis. Individual HvCBF genes differed in response to abiotic stress types and in the response time frame, suggesting different sets of HvCBF genes are employed relative to particular stresses. HvCBFs specifically bound monocot and dicot cor gene CRT elements in vitro under both warm and cold conditions; however, binding of HvCBF4-subgroup members was cold dependent. The temperature-independent HvCBFs activated cor gene expression at warm temperatures in transgenic Arabidopsis, while the cold-dependent HvCBF4-subgroup members of three Triticeae species did not. These results suggest that in the Triticeae - as in Arabidopsis - members of the CBF gene family function as fundamental components of the winter hardiness regulon.

Molecular and structural characterization of barley vernalization genes.

Vernalization, the requirement of a period of low temperature to induce transition from the vegetative to reproductive state, is an evolutionarily and economically important trait in the Triticeae. The genetic basis of vernalization in cultivated barley (Hordeum vulgare subsp. vulgare) can be defined using the two-locus VRN-H1/VRN-H2 model. We analyzed the allelic characteristics of HvBM5A, the candidate gene for VRN-H1, from ten cultivated barley accessions and one wild progenitor accession (subsp. spontaneum), representing the three barley growth habits - winter, facultative, and spring. We present multiple lines of evidence, including sequence, linkage map location, and expression, that support HvBM5A being VRN-H1. While the predicted polypeptides from different growth habits are identical, spring accessions contain a deletion in the first intron of HvBM5A that may be important for regulation. While spring HvBM5A alleles are typified by the intron-localized deletion, in some cases, the promoter may also determine the allele type. The presence/absence of the tightly linked ZCCT-H gene family members on chromosome 4H perfectly correlates with growth habit and we conclude that one of the three ZCCT-H genes is VRN-H2. The VRN-H2 locus is present in winter genotypes and deleted from the facultative and spring genotypes analyzed in this study, suggesting the facultative growth habit (cold tolerant, vernalization unresponsive) is a result of deletion of the VRN-H2 locus and presence of a winter HvBM5A allele. All reported barley vernalization QTLs can be explained by the two-locus VRN-H1/VRN-H2 model based on the presence/absence of VRN-H2 and a winter vs. spring HvBM5A allele.

Large deletions within the first intron in VRN-1 are associated with spring growth habit in barley and wheat.

The broad adaptability of wheat and barley is in part attributable to their flexible growth habit, in that spring forms have recurrently evolved from the ancestral winter growth habit. In diploid wheat and barley growth habit is determined by allelic variation at the VRN-1 and/or VRN-2 loci, whereas in the polyploid wheat species it is determined primarily by allelic variation at VRN-1. Dominant Vrn-A1 alleles for spring growth habit are frequently associated with mutations in the promoter region in diploid wheat and in the A genome of common wheat. However, several dominant Vrn-A1, Vrn-B1, Vrn-D1 (common wheat) and Vrn-H1 (barley) alleles show no polymorphisms in the promoter region relative to their respective recessive alleles. In this study, we sequenced the complete VRN-1 gene from these accessions and found that all of them have large deletions within the first intron, which overlap in a 4-kb region. Furthermore, a 2.8-kb segment within the 4-kb region showed high sequence conservation among the different recessive alleles. PCR markers for these deletions showed that similar deletions were present in all the accessions with known Vrn-B1 and Vrn-D1 alleles, and in 51 hexaploid spring wheat accessions previously shown to have no polymorphisms in the VRN-A1 promoter region. Twenty-four tetraploid wheat accessions had a similar deletion in VRN-A1 intron 1. We hypothesize that the 2.8-kb conserved region includes regulatory elements important for the vernalization requirement. Epistatic interactions between VRN-H2 and the VRN-H1 allele with the intron 1 deletion suggest that the deleted region may include a recognition site for the flowering repression mediated by the product of the VRN-H2 gene of barley.

Molecular characterization of the duplicated meristem identity genes HvAP1a and HvAP1b in barley.

The vernalization gene VRN-1 has been identified as a MADS-box transcription factor orthologous to the meristem identity gene APETALA1 (AP1). A single copy of this gene was found in diploid wheat, but 2 copies were reported in barley. In this study, we present a detailed characterization of these 2 copies to understand their respective roles in the vernalization response. We identified 2 groups of barley bacterial artificial chromosomes (BACs), each containing 1 AP1 copy designated hereafter as HvAP1a and HvAP1b. A physical map of the VRN-H1 region showed that the HvAP1a BACs were part of the VRN-H1 region but that the HvAP1b BACs were not. Numerous structural changes were observed between the barley and wheat VRN-1 physical maps. In a population segregating for VRN-H1, the HvAP1a gene cosegregated with growth habit, suggesting that HvAP1a is the barley vernalization gene VRN-H1. The other copy, HvAP1b, was mapped on the centromeric region of chromosome 1H, the chromosome where vernalization gene VRN-H3 was previously mapped. We developed a mapping population segregating for VRN-H3 and showed that 2 molecular makers flanking HvAP1b locus were not linked to growth habit. The HvAP1b copy has a complete deletion of the first 2 exons, suggesting that it is a truncated pseudogene and not a candidate for VRN-H3. In summary, this study contributed a detailed physical map of the barley VRN-H1 region, showed several structural differences with the orthologous wheat region, and clarified the identity of the barley VRN-H1 gene.