Haplotype‐based genotyping‐by‐sequencing in oat genome research

In a de novo genotyping‐by‐sequencing (GBS) analysis of short, 64‐base tag‐level haplotypes in 4657 accessions of cultivated oat, we discovered 164741 tag‐level (TL) genetic variants containing 241224 SNPs. From this, the marker density of an oat consensus map was increased by the addition of more than 70000 loci. The mapped TL genotypes of a 635‐line diversity panel were used to infer chromosome‐level (CL) haplotype maps. These maps revealed differences in the number and size of haplotype blocks, as well as differences in haplotype diversity between chromosomes and subsets of the diversity panel. We then explored potential benefits of SNP vs. TL vs. CL GBS variants for mapping, high‐resolution genome analysis and genomic selection in oats. A combined genome‐wide association study (GWAS) of heading date from multiple locations using both TL haplotypes and individual SNP markers identified 184 significant associations. A comparative GWAS using TL haplotypes, CL haplotype blocks and their combinations demonstrated the superiority of using TL haplotype markers. Using a principal component‐based genome‐wide scan, genomic regions containing signatures of selection were identified. These regions may contain genes that are responsible for the local adaptation of oats to Northern American conditions. Genomic selection for heading date using TL haplotypes or SNP markers gave comparable and promising prediction accuracies of up to r = 0.74. Genomic selection carried out in an independent calibration and test population for heading date gave promising prediction accuracies that ranged between r = 0.42 and 0.67. In conclusion, TL haplotype GBS‐derived markers facilitate genome analysis and genomic selection in oat.     

Increasing the accuracy of genomic prediction in pure‐bred Limousin beef cattle by including cross‐bred Limousin data and accounting for an F94L variant in MSTN

Explicitly fitting effects for major genes or QTL that account for a large percentage of variation in a whole genomic prediction model may increase prediction accuracy. This study compared approaches to account for a major effect of an F94L variant in the MSTN gene within the genomic prediction using bovine whole‐genomic SNP markers. Among the beef cattle breeds, Limousin have been known to have an F94L variant that is not present in Angus. The reference population in this study consisted of 3060 beef cattle including pure‐bred Limousin (PL), cross‐bred Limousin with Angus (LF) and pure‐bred Angus, genotyped using a BovineSNP50 BeadChip and directly for the MSTN‐F94L variant. We compared prediction accuracies in PL animals using the three datasets from only the PL population, admixed PL and LF (AL) or multibreed analysis using all of the PL, LF and Angus (MB) population according to four‐fold cross‐validation after K‐means clustering. The MSTN‐F94L variant was the most strongly associated with five traits (birth weight, calving ease direct, milk, weaning weight and yield grade) among the 13 measured traits in PL and AL populations. Fitting the MSTN‐F94L variant as a random effect, the genomic prediction accuracies for birth weight increased by 2.7% in PL, by 2.2% in AL and by 3.2% in MB. Prediction accuracies for five traits increased in the MB analysis. Fitting MSTN‐F94L as a fixed effect in PL, AL and MB analyses resulted in increased prediction accuracy in PL for eight traits. Prediction accuracies can be improved by including a causal variant in genomic evaluation compared with simply using whole‐genome SNP markers. Fitting the causal variant as a fixed effect along with markers fitted as random effects resulted in greater prediction accuracies for most traits. Causal variants should be genotyped along with SNP markers.     

A comprehensive draft genome sequence for lupin (Lupinus angustifolius), an emerging health food: insights into plant–microbe interactions and legume evolution

Lupins are important grain legume crops that form a critical part of sustainable farming systems, reducing fertilizer use and providing disease breaks. It has a basal phylogenetic position relative to other crop and model legumes and a high speciation rate. Narrow‐leafed lupin (NLL; Lupinus angustifolius L.) is gaining popularity as a health food, which is high in protein and dietary fibre but low in starch and gluten‐free. We report the draft genome assembly (609 Mb) of NLL cultivar Tanjil, which has captured >98% of the gene content, sequences of additional lines and a dense genetic map. Lupins are unique among legumes and differ from most other land plants in that they do not form mycorrhizal associations. Remarkably, we find that NLL has lost all mycorrhiza‐specific genes, but has retained genes commonly required for mycorrhization and nodulation. In addition, the genome also provided candidate genes for key disease resistance and domestication traits. We also find evidence of a whole‐genome triplication at around 25 million years ago in the genistoid lineage leading to Lupinus. Our results will support detailed studies of legume evolution and accelerate lupin breeding programmes.     

High‐throughput genomics in sorghum: from whole‐genome resequencing to a SNP screening array

With its small, diploid and completely sequenced genome, sorghum (Sorghum bicolor L. Moench) is highly amenable to genomics‐based breeding approaches. Here, we describe the development and testing of a robust single‐nucleotide polymorphism (SNP) array platform that enables polymorphism screening for genome‐wide and trait‐linked polymorphisms in genetically diverse S. bicolor populations. Whole‐genome sequences with 6× to 12× coverage from five genetically diverse S. bicolor genotypes, including three sweet sorghums and two grain sorghums, were aligned to the sorghum reference genome. From over 1 million high‐quality SNPs, we selected 2124 Infinium Type II SNPs that were informative in all six source genomes, gave an optimal Assay Design Tool (ADT) score, had allele frequencies of 50% in the six genotypes and were evenly spaced throughout the S. bicolor genome. Furthermore, by phenotype‐based pool sequencing, we selected an additional 876 SNPs with a phenotypic association to early‐stage chilling tolerance, a key trait for European sorghum breeding. The 3000 attempted bead types were used to populate half of a dual‐species Illumina iSelect SNP array. The array was tested using 564 Sorghum spp. genotypes, including offspring from four unrelated recombinant inbred line (RIL) and F₂ populations and a genetic diversity collection. A high call rate of over 80% enabled validation of 2620 robust and polymorphic sorghum SNPs, underlining the efficiency of the array development scheme for whole‐genome SNP selection and screening, with diverse applications including genetic mapping, genome‐wide association studies and genomic selection.     

SNP discovery in nonmodel organisms: strand bias and base‐substitution errors reduce conversion rates

Single nucleotide polymorphisms (SNPs) have become the marker of choice for genetic studies in organisms of conservation, commercial or biological interest. Most SNP discovery projects in nonmodel organisms apply a strategy for identifying putative SNPs based on filtering rules that account for random sequencing errors. Here, we analyse data used to develop 4723 novel SNPs for the commercially important deep‐sea fish, orange roughy (Hoplostethus atlanticus), to assess the impact of not accounting for systematic sequencing errors when filtering identified polymorphisms when discovering SNPs. We used SAMtools to identify polymorphisms in a velvet assembly of genomic DNA sequence data from seven individuals. The resulting set of polymorphisms were filtered to minimize ‘bycatch’—polymorphisms caused by sequencing or assembly error. An Illumina Infinium SNP chip was used to genotype a final set of 7714 polymorphisms across 1734 individuals. Five predictors were examined for their effect on the probability of obtaining an assayable SNP: depth of coverage, number of reads that support a variant, polymorphism type (e.g. A/C), strand‐bias and Illumina SNP probe design score. Our results indicate that filtering out systematic sequencing errors could substantially improve the efficiency of SNP discovery. We show that BLASTX can be used as an efficient tool to identify single‐copy genomic regions in the absence of a reference genome. The results have implications for research aiming to identify assayable SNPs and build SNP genotyping assays for nonmodel organisms.     

Cross‐platform compatibility of de novo‐aligned SNPs in a nonmodel butterfly genus

High‐throughput sequencing methods for genotyping genome‐wide markers are being rapidly adopted for phylogenetics of nonmodel organisms in conservation and biodiversity studies. However, the reproducibility of SNP genotyping and degree of marker overlap or compatibility between datasets from different methodologies have not been tested in nonmodel systems. Using double‐digest restriction site‐associated DNA sequencing, we sequenced a common set of 22 specimens from the butterfly genus Speyeria on two different Illumina platforms, using two variations of library preparation. We then used a de novo approach to bioinformatic locus assembly and SNP discovery for subsequent phylogenetic analyses. We found a high rate of locus recovery despite differences in library preparation and sequencing platforms, as well as overall high levels of data compatibility after data processing and filtering. These results provide the first application of NGS methods for phylogenetic reconstruction in Speyeria and support the use and long‐term viability of SNP genotyping applications in nonmodel systems.     

A new set of polymorphic simple sequence repeat (SSR) markers from a wild strawberry (Fragaria vesca) are transferable to other diploid Fragaria species and to Fragaria × ananassa

A set of 41 polymorphic microsatellite markers were developed using a CT/AG‐enriched genomic library of Fragaria vesca cv. Reine des Vallées. Thirty‐five of them were polymorphic in F. vesca and were tested in one accession each of six additional diploid Fragaria species and the octoploid Fragaria× ananassa. A mean of 5.3 alleles per locus and a low level of observed heterozygosity were generally detected in the 32 single‐locus simple sequence repeats of F. vesca. Most of these loci amplify in the other diploid species and in F. × ananassa.     

Whole‐genome resequencing‐based QTL‐seq identified candidate genes and molecular markers for fresh seed dormancy in groundnut

The subspecies fastigiata of cultivated groundnut lost fresh seed dormancy (FSD) during domestication and human‐made selection. Groundnut varieties lacking FSD experience precocious seed germination during harvest imposing severe losses. Development of easy‐to‐use genetic markers enables early‐generation selection in different molecular breeding approaches. In this context, one recombinant inbred lines (RIL) population (ICGV 00350 × ICGV 97045) segregating for FSD was used for deploying QTL‐seq approach for identification of key genomic regions and candidate genes. Whole‐genome sequencing (WGS) data (87.93 Gbp) were generated and analysed for the dormant parent (ICGV 97045) and two DNA pools (dormant and nondormant). After analysis of resequenced data from the pooled samples with dormant parent (reference genome), we calculated delta‐SNP index and identified a total of 10,759 genomewide high‐confidence SNPs. Two candidate genomic regions spanning 2.4 Mb and 0.74 Mb on the B05 and A09 pseudomolecules, respectively, were identified controlling FSD. Two candidate genes—RING‐H2 finger protein and zeaxanthin epoxidase—were identified in these two regions, which significantly express during seed development and control abscisic acid (ABA) accumulation. QTL‐seq study presented here laid out development of a marker, GMFSD1, which was validated on a diverse panel and could be used in molecular breeding to improve dormancy in groundnut.     

QTL‐seq approach identified genomic regions and diagnostic markers for rust and late leaf spot resistance in groundnut (Arachis hypogaea L.)

Rust and late leaf spot (LLS) are the two major foliar fungal diseases in groundnut, and their co‐occurrence leads to significant yield loss in addition to the deterioration of fodder quality. To identify candidate genomic regions controlling resistance to rust and LLS, whole‐genome resequencing (WGRS)‐based approach referred as ‘QTL‐seq’ was deployed. A total of 231.67 Gb raw and 192.10 Gb of clean sequence data were generated through WGRS of resistant parent and the resistant and susceptible bulks for rust and LLS. Sequence analysis of bulks for rust and LLS with reference‐guided resistant parent assembly identified 3136 single‐nucleotide polymorphisms (SNPs) for rust and 66 SNPs for LLS with the read depth of ≥7 in the identified genomic region on pseudomolecule A03. Detailed analysis identified 30 nonsynonymous SNPs affecting 25 candidate genes for rust resistance, while 14 intronic and three synonymous SNPs affecting nine candidate genes for LLS resistance. Subsequently, allele‐specific diagnostic markers were identified for three SNPs for rust resistance and one SNP for LLS resistance. Genotyping of one RIL population (TAG 24 × GPBD 4) with these four diagnostic markers revealed higher phenotypic variation for these two diseases. These results suggest usefulness of QTL‐seq approach in precise and rapid identification of candidate genomic regions and development of diagnostic markers for breeding applications.     

Development of Diversity Arrays Technology markers as a tool for rapid genomic assessment in Nile tilapia,Oreochromis niloticus

The development of genomic markers is described for Nile tilapia, Oreochromis niloticus, using the Diversity Arrays Technology (DArT) genotype‐by‐sequencing platform. A total of 13 215 single nucleotide polymorphism (SNP) markers and 12 490 silicoDArT (dominant) markers were identified from broodstock of two selective breeding programs [Genetically Improved Farmed Tilapia (GIFT) strain from Malaysia and the Abbassa strain from Egypt]. Over 10 000 SNPs were polymorphic in either strain, and 2985 and 3087 showed strain‐specific polymorphisms for the GIFT and Abbassa strains respectively. We demonstrate the potential utility of these markers for rapid genomic screening and use in breeding programs.     

Single‐nucleotide polymorphism discovery and diversity in the model legume Medicago truncatula

Extensive genomic resources are available in the model legume Medicago truncatula. Here, we present the discovery and design of the first array of single‐nucleotide polymorphism (SNP) markers in M. truncatula through large‐scale Sanger resequencing of genomic fragments spanning the genome, in a diverse panel of 16 M. truncatula accessions. Both anonymous fragments and fragments targeting candidate genes for flowering phenology and symbiosis were surveyed for nucleotide variation in almost 230 kb of unique genomic regions. A set of 384 SNP markers was designed for an Illumina's GoldenGate assay, genotyped on a collection of 192 inbred lines (CC192) representing the geographical range of the species and used to survey the diversity of two natural populations. Finally, 86% of the tested SNPs were of high quality and exhibited polymorphism in the CC192 collection. Even at the population level, we detected polymorphism for more than 50% of the selected SNPs. Analysis of the allele frequency spectrum in the CC192 showed a reduced ascertainment bias, mostly limited to very rare alleles (frequency <0.01). The substantial polymorphism detected at the species and population levels, the high marker quality and the potential to survey large samples of individuals make this set of SNP markers a valuable tool to improve our understanding of the effect of demographic and selective factors that shape the natural genetic diversity within the selfing species Medicago truncatula.     

Genetic dissection of pre‐anthesis sub‐phase durations during the reproductive spike development of wheat

Flowering time is an important factor affecting grain yield in wheat. In this study, we divided reproductive spike development into eight sub‐phases. These sub‐phases have the potential to be delicately manipulated to increase grain yield. We measured 36 traits with regard to sub‐phase durations, determined three grain yield‐related traits in eight field environments and mapped 15 696 single nucleotide polymorphism (SNP, based on 90k Infinium chip and 35k Affymetrix chip) markers in 210 wheat genotypes. Phenotypic and genetic associations between grain yield traits and sub‐phase durations showed significant consistency (Mantel test; r = 0.5377, P < 0.001). The shared quantitative trait loci (QTLs) revealed by the genome‐wide association study suggested a close association between grain yield and sub‐phase duration, which may be attributed to effects on spikelet initiation/spikelet number (double ridge to terminal spikelet stage, DR‐TS) and assimilate accumulation (green anther to anthesis stage, GA‐AN). Moreover, we observed that the photoperiod‐sensitivity allele at the Ppd‐D1 locus on chromosome 2D markedly extended all sub‐phase durations, which may contribute to its positive effects on grain yield traits. The dwarfing allele at the Rht‐D1 (chromosome 4D) locus altered the sub‐phase duration and displayed positive effects on grain yield traits. Data for 30 selected genotypes (from among the original 210 genotypes) in the field displayed a close association with that from the greenhouse. Most importantly, this study demonstrated specific connections to grain yield in narrower time windows (i.e. the eight sub‐phases), rather than the entire stem elongation phase as a whole.