Stig U. Andersen

In 2008 we published the first set of guidelines for standardizing research in autophagy. Since then, research on this topic has continued to accelerate, and many new scientists have entered the field. Our knowledge base and relevant new technologies have also been expanding. Accordingly, it is important to update these guidelines for monitoring autophagy in different organisms. Various reviews have described the range of assays that have been used for this purpose. Nevertheless, there continues to be confusion regarding acceptable methods to measure autophagy, especially in multicellular eukaryotes. For example, a key point that needs to be emphasized is that there is a difference between measurements that monitor the numbers or volume of autophagic elements (e.g., autophagosomes or autolysosomes) at any stage of the autophagic process versus those that measure flux through the autophagy pathway (i.e., the complete process including the amount and rate of cargo sequestered and degraded). In particular, a block in macroautophagy that results in autophagosome accumulation must be differentiated from stimuli that increase autophagic activity, defined as increased autophagy induction coupled with increased delivery to, and degradation within, lysosomes (in most higher eukaryotes and some protists such as Dictyostelium) or the vacuole (in plants and fungi). In other words, it is especially important that investigators new to the field understand that the appearance of more autophagosomes does not necessarily equate with more autophagy. In fact, in many cases, autophagosomes accumulate because of a block in trafficking to lysosomes without a concomitant change in autophagosome biogenesis, whereas an increase in autolysosomes may reflect a reduction in degradative activity. It is worth emphasizing here that lysosomal digestion is a stage of autophagy and evaluating its competence is a crucial part of the evaluation of autophagic flux, or complete autophagy. Here, we present a set of guidelines for the selection and interpretation of methods for use by investigators who aim to examine macroautophagy and related processes, as well as for reviewers who need to provide realistic and reasonable critiques of papers that are focused on these processes. These guidelines are not meant to be a formulaic set of rules, because the appropriate assays depend in part on the question being asked and the system being used. In addition, we emphasize that no individual assay is guaranteed to be the most appropriate one in every situation, and we strongly recommend the use of multiple assays to monitor autophagy. Along these lines, because of the potential for pleiotropic effects due to blocking autophagy through genetic manipulation, it is imperative to target by gene knockout or RNA interference more than one autophagy-related protein. In addition, some individual Atg proteins, or groups of proteins, are involved in other cellular pathways implying that not all Atg proteins can be used as a specific marker for an autophagic process. In these guidelines, we consider these various methods of assessing autophagy and what information can, or cannot, be obtained from them. Finally, by discussing the merits and limits of particular assays, we hope to encourage technical innovation in the field.

In 2008, we published the first set of guidelines for standardizing research in autophagy. Since then, this topic has received increasing attention, and many scientists have entered the field. Our knowledge base and relevant new technologies have also been expanding. Thus, it is important to formulate on a regular basis updated guidelines for monitoring autophagy in different organisms. Despite numerous reviews, there continues to be confusion regarding acceptable methods to evaluate autophagy, especially in multicellular eukaryotes. Here, we present a set of guidelines for investigators to select and interpret methods to examine autophagy and related processes, and for reviewers to provide realistic and reasonable critiques of reports that are focused on these processes. These guidelines are not meant to be a dogmatic set of rules, because the appropriateness of any assay largely depends on the question being asked and the system being used. Moreover, no individual assay is perfect for every situation, calling for the use of multiple techniques to properly monitor autophagy in each experimental setting. Finally, several core components of the autophagy machinery have been implicated in distinct autophagic processes (canonical and noncanonical autophagy), implying that genetic approaches to block autophagy should rely on targeting two or more autophagy-related genes that ideally participate in distinct steps of the pathway. Along similar lines, because multiple proteins involved in autophagy also regulate other cellular pathways including apoptosis, not all of them can be used as a specific marker for bona fide autophagic responses. Here, we critically discuss current methods of assessing autophagy and the information they can, or cannot, provide. Our ultimate goal is to encourage intellectual and technical innovation in the field.

Faba bean (Vicia faba L.) seeds are an excellent source of plant-based protein. In spite of the vast nutritional and environmental benefits provided by faba bean cultivation, its use as a food crop has been restricted, primarily due to the presence of the pyrimidine glycosides vicine and convicine (v-c). Ingestion of v-c can cause favism in individuals with a genetically inherited deficiency in glucose-6-phosphate dehydrogenase (G6PD). In monogastric animals, v-c can cause decreased feeding efficiency. The elimination of these glucosides is a goal of most faba bean breeding programs worldwide. Our review focuses on the current genetic, molecular and biochemical knowledge surrounding the accumulation of v-c in faba beans. The gap between the current knowledge and what remains unknown is discussed. This review also explores historical and obscure information on v-c in faba bean. A low-v-c faba bean line was identified in the 1980s and this trait has been introduced into several modern cultivars. It has been shown that low-v-c faba beans are safe for G6PD-deficient individuals. A robust molecular marker is now available for marker-assisted breeding to reduce levels of v-c. The biosynthetic pathway of v-c is not yet understood and is currently under investigation. An international coordinated effort, led by the authors of this paper, is making progress towards full elucidation of the pathway. Further efforts in this direction could lead to lower levels of these compounds than the current low v-c genotypes offer, perhaps even complete elimination.

Biological nitrogen fixation by Rhizobium-legume symbioses represents an environmentally friendly and inexpensive alternative to the use of chemical nitrogen fertilizers in legume crops. Rhizobial inoculants, applied frequently as biofertilizers, play an important role in sustainable agriculture. However, inoculants often fail to compete for nodule occupancy against native rhizobia with inferior nitrogen-fixing abilities, resulting in low yields. Strains with excellent performance under controlled conditions are typically selected as inoculants, but the rates of nodule occupancy compared to native strains are rarely investigated. Lack of persistence in the field after agricultural cycles, usually due to the transfer of symbiotic genes from the inoculant strain to naturalized populations, also limits the suitability of commercial inoculants. When rhizobial inoculants are based on native strains with a high nitrogen fixation ability, they often have superior performance in the field due to their genetic adaptations to the local environment. Therefore, knowledge from laboratory studies assessing competition and understanding how diverse strains of rhizobia behave, together with assays done under field conditions, may allow us to exploit the effectiveness of native populations selected as elite strains and to breed specific host cultivar-rhizobial strain combinations. Here, we review current knowledge at the molecular level on competition for nodulation and the advances in molecular tools for assessing competitiveness. We then describe ongoing approaches for inoculant development based on native strains and emphasize future perspectives and applications using a multidisciplinary approach to ensure optimal performance of both symbiotic partners.

Faba bean (Vicia faba L.), a member of the Fabaceae family, is one of the important food legumes cultivated in cool temperate regions. It holds great importance for human consumption and livestock feed because of its high protein content, dietary fibre, and nutritional value. Major faba bean breeding challenges include its mixed breeding system, unknown wild progenitor, and genome size of ~13 Gb, which is the largest among diploid field crops. The key breeding objectives in faba bean include improved resistance to biotic and abiotic stress and enhanced seed quality traits. Regarding quality traits, major progress on reduction of vicine-convicine and seed coat tannins, the main anti-nutritional factors limiting faba bean seed usage, have been recently achieved through gene discovery. Genomic resources are relatively less advanced compared with other grain legume species, but significant improvements are underway due to a recent increase in research activities. A number of bi-parental populations have been constructed and mapped for targeted traits in the last decade. Faba bean now benefits from saturated synteny-based genetic maps, along with next-generation sequencing and high-throughput genotyping technologies that are paving the way for marker-assisted selection. Developing a reference genome, and ultimately a pan-genome, will provide a foundational resource for molecular breeding. In this review, we cover the recent development and deployment of genomic tools for faba bean breeding.

The plant-associated microbiome is a key component of plant systems, contributing to their health, growth, and productivity. The application of machine learning (ML) in this field promises to help untangle the relationships involved. However, measurements of microbial communities by high-throughput sequencing pose challenges for ML. Noise from low sample sizes, soil heterogeneity, and technical factors can impact the performance of ML. Additionally, the compositional and sparse nature of these datasets can impact the predictive accuracy of ML. We review recent literature from plant studies to illustrate that these properties often go unmentioned. We expand our analysis to other fields to quantify the degree to which mitigation approaches improve the performance of ML and describe the mathematical basis for this. With the advent of accessible analytical packages for microbiome data including learning models, researchers must be familiar with the nature of their datasets.

Hydroxyurea (HU) is a well tolerated ribonucleotide reductase inhibitor effective in HIV, sickle cell disease, and blood cancer therapy. Despite a positive initial response, however, most treated cancers eventually progress due to development of HU resistance. Although RNR properties influence HU resistance in cell lines, the mechanisms underlying cancer HU resistance in vivo remain unclear. To address this issue, we screened for HU resistance in the plant Arabidopsis thaliana and identified seventeen unique catalase mutants, thereby establishing that HU toxicity depends on catalase in vivo. We further demonstrated that catalase is a direct HU target by showing that HU acts as a competitive inhibitor of catalase-mediated hydrogen peroxide decomposition. Considering also that catalase can accelerate HU decomposition in vitro and that co-treatment with another catalase inhibitor alleviates HU effects in vivo, our findings suggests that HU could act as a catalase-activated pro-drug. Clinically, we found high catalase activity in circulating cells from untreated chronic myeloid leukemia, offering a possible explanation for the efficacy of HU against this malignancy.

Morphogens provide positional information and their concentration is key to the organized development of multicellular organisms. Nitrogen-fixing root nodules are unique organs induced by Nod factor-producing bacteria. Localized production of Nod factors establishes a developmental field within the root where plant cells are reprogrammed to form infection threads and primordia. We found that regulation of Nod factor levels by Lotus japonicus is required for the formation of nitrogen-fixing organs, determining the fate of this induced developmental program. Our analysis of plant and bacterial mutants shows that a host chitinase modulates Nod factor levels possibly in a structure-dependent manner. In Lotus, this is required for maintaining Nod factor signalling in parallel with the elongation of infection threads within the nodule cortex, while root hair infection and primordia formation are not influenced. Our study shows that infected nodules require balanced levels of Nod factors for completing their transition to functional, nitrogen-fixing organs.

Legume symbiosis with rhizobia results in the formation of a specialized organ, the root nodule, where atmospheric dinitrogen is reduced to ammonia. In Lotus japonicus ( Lotus ), several genes involved in nodule development or nodule function have been defined using biochemistry, genetic approaches, and high‐throughput transcriptomics. We have employed proteomics to further understand nodule development. Two developmental stages representing nodules prior to nitrogen fixation (white) and mature nitrogen fixing nodules (red) were compared with roots. In addition, the proteome of a spontaneous nodule formation mutant ( snf1 ) was determined. From nodules and roots, 780 and 790 protein spots from 2D gels were identified and approximately 45% of the corresponding unique gene accessions were common. Including a previous proteomics set from Lotus pod and seed, the common gene accessions were decreased to 7%. Interestingly, an indication of more pronounced PTMs in nodules than in roots was determined. Between the two nodule developmental stages, higher levels of pathogen‐related 10 proteins, HSPs, and proteins involved in redox processes were found in white nodules, suggesting a higher stress level at this developmental stage. In contrast, protein spots corresponding to nodulins such as leghemoglobin, asparagine synthetase, sucrose synthase, and glutamine synthetase were prevalent in red nodules. The distinct biochemical state of nodules was further highlighted by the conspicuous presence of several nitrilases, ascorbate metabolic enzymes, and putative rhizobial effectors.

Abstract The biological interactions between plants and their root microbiomes are essential for plant growth, and even though plant genotype [G], soil microbiome [M], and growth conditions (environment) [E] are the core factors shaping root microbiome, their relationships remain unclear. In this study we investigated the effects of G, M, and E and their interactions on the Lotus root microbiome and plant growth using an in vitro cross-inoculation approach which reconstructed the interactions between nine Lotus accessions and four soil microbiomes under two different environmental conditions. Results suggested that a large proportion of the root microbiome composition is determined by M and E, while G-related (G, G × M, and G × E) effects were significant but small. In contrast, the interaction between G and M had a more pronounced effect on plant shoot growth than M alone. Our findings also indicated that most microbiome variations controlled by M have little effect on plant phenotypes, whereas G × M interactions have more significant effects. Plant genotype-dependent interactions with soil microbes warrant more attention to optimize crop yield and resilience.

Abstract Generation of doubled haploid plants is a powerful tool in breeding, as homozygous individuals will be obtained directly from hybrids. However, genotype variability in regeneration efficiency of most European wheat ( Triticum aestivum L.) varieties has limited its use in wheat. This study intended to identify quantitative trait loci ( QTL s) for green plantlet regeneration from wheat microspore cultures. A QTL analysis using DA r T markers was conducted based on a bi‐parental F 3 population, derived from a cross between the varieties Svilena and Jensen, which displayed markedly different capacity for plantlet regeneration. Two QTL s on chromosome 1B and 7B explained 53% of the variation in green plantlet regeneration. Furthermore, a collection of 94 European wheat varieties was genotyped and phenotyped. The microspore response level was low among western and northern European wheat varieties, and the positive QTL s found in the bi‐parental population were rare in the variety collection. Identification of the two QTL s enables introduction of high regeneration efficiency into wheat germplasm. Moreover, our results proved that the efficient regeneration observed for one variety could be crossed into modern winter wheat.

EDITORIAL article Front. Plant Sci., 12 December 2018Sec. Plant Pathogen Interactions Volume 9 - 2018 | https://doi.org/10.3389/fpls.2018.01839

Homologous recombination is expected to increase natural selection efficacy by decoupling the fate of beneficial and deleterious mutations and by readily creating new combinations of beneficial alleles. Here, we investigate how the proportion of amino acid substitutions fixed by adaptive evolution (α) depends on the recombination rate in bacteria. We analyze 3,086 core protein-coding sequences from 196 genomes belonging to five closely related species of the genus Rhizobium. These genes are found in all species and do not display any signs of introgression between species. We estimate α using the site frequency spectrum (SFS) and divergence data for all pairs of species. We evaluate the impact of recombination within each species by dividing genes into three equally sized recombination classes based on their average level of intragenic linkage disequilibrium. We find that α varies from 0.07 to 0.39 across species and is positively correlated with the level of recombination. This is both due to a higher estimated rate of adaptive evolution and a lower estimated rate of nonadaptive evolution, suggesting that recombination both increases the fixation probability of advantageous variants and decreases the probability of fixation of deleterious variants. Our results demonstrate that homologous recombination facilitates adaptive evolution measured by α in the core genome of prokaryote species in agreement with studies in eukaryotes.

Recombinant inbred lines (RILs) derived from bi-parental populations are stable genetic resources, which are widely used for constructing genetic linkage maps. These genetic maps are essential for QTL mapping and can aid contig and scaffold anchoring in the final stages of genome assembly. In this study, two Lotus sp. RIL populations, Lotus japonicus MG-20 × Gifu and Gifu × L. burttii, were characterized by Illumina re-sequencing. Genotyping of 187 MG-20 × Gifu RILs at 87,140 marker positions and 96 Gifu × L. burttii RILs at 357,973 marker positions allowed us to accurately identify 1,929 recombination breakpoints in the MG-20 × Gifu RILs and 1,044 breakpoints in the Gifu × L. burttii population. The resulting high-density genetic maps now facilitate high-accuracy QTL mapping, identification of reference genome mis-assemblies, and characterization of structural variants.

Forward and reverse genetics using the model legumes Lotus japonicus and Medicago truncatula have been instrumental in identifying the essential genes governing legume-rhizobia symbiosis. However, little information is known about the effects of intraspecific variation on symbiotic signalling. Here, we use quantitative trait locus sequencing (QTL-seq) to investigate the genetic basis of the differentiated phenotypic responses shown by the Lotus accessions Gifu and MG20 to inoculation with the Mesorhizobium loti exoU mutant that produces truncated exopolysaccharides. We identified through genetic complementation the Pxy gene as a component of this differential exoU response. Lotus Pxy encodes a leucine-rich repeat receptor-like kinase similar to Arabidopsis thaliana PXY, which regulates stem vascular development. We show that Lotus pxy insertion mutants displayed defects in root and stem vascular organisation, as well as lateral root and nodule formation. Our work links Pxy to de novo organogenesis in the root, highlights the genetic overlap between regulation of lateral root and nodule formation, and demonstrates that natural variation in Pxy affects nodulation signalling.

Abstract While shaping of plant microbiome composition through ‘host filtering’ is well documented in legume–rhizobium symbioses, it is less clear to what extent different varieties and genotypes of the same plant species differentially influence symbiont community diversity and composition. Here, we compared how clover host varieties and genotypes affect the structure of Rhizobium populations in root nodules under conventional field and controlled greenhouse conditions. We first grew four Trifolium repens (white clover) F 2 crosses and one variety in a conventional field trial and compared differences in root nodule Rhizobium leguminosarum symbiovar trifolii (Rlt) genotype diversity using high‐throughput amplicon sequencing of chromosomal housekeeping ( rpoB and recA ) genes and auxiliary plasmid‐borne symbiosis genes ( nodA and nodD ). We found that Rlt nodule diversities significantly differed between clover crosses, potentially due to host filtering. However, variance in Rlt diversity largely overlapped between crosses and was also explained by the spatial distribution of plants in the field, indicative of the role of local environmental conditions for nodule diversity. To test the effect of host filtering, we conducted a controlled greenhouse trial with a diverse Rlt inoculum and several host genotypes. We found that different clover varieties and genotypes of the same variety selected for significantly different Rlt nodule communities and that the strength of host filtering (deviation from the initial Rhizobium inoculant composition) was positively correlated with the efficiency of symbiosis (rate of plant greenness colouration). Together, our results suggest that selection by host genotype and local growth conditions jointly influence white clover Rlt nodule diversity and community composition.

Legumes acquire access to atmospheric nitrogen through nitrogen fixation by rhizobia in root nodules. Rhizobia are soil-dwelling bacteria and there is a tremendous diversity of rhizobial species in different habitats. From the legume perspective, host range is a compromise between the ability to colonize new habitats, in which the preferred symbiotic partner may be absent, and guarding against infection by suboptimal nitrogen fixers. Here, we investigate natural variation in rhizobial host range across Lotus species. We find that Lotus burttii is considerably more promiscuous than Lotus japonicus, represented by the Gifu accession, in its interactions with rhizobia. This promiscuity allows Lotus burttii to form nodules with Mesorhizobium, Rhizobium, Sinorhizobium, Bradyrhizobium, and Allorhizobium species that represent five distinct genera. Using recombinant inbred lines, we have mapped the Gifu/burttii promiscuity quantitative trait loci (QTL) to the same genetic locus regardless of rhizobial genus, suggesting a general genetic mechanism for symbiont-range expansion. The Gifu/burttii QTL now provides an opportunity for genetic and mechanistic understanding of promiscuous legume-rhizobia interactions. [Formula: see text] Copyright © 2022 The Author(s). This is an open access article distributed under the CC BY 4.0 International license.

Nature Communications 5: Article number: 3765 (2014); Published: 6 May 2014; Updated: 6 August 2014. The original version of the Supplementary Information attached to this Article contained an error in the numbering of the Supplementary Figures and Tables. The HTML has now been updated to include a corrected version of the Supplementary Information.

Abstract Nucleotide-binding site leucine-rich repeat resistance genes (NLRs) allow plants to detect microbial effectors. We hypothesized that NLR expression patterns would reflect organ-specific differences in effector challenge and tested this by carrying out a meta-analysis of expression data for 1,235 NLRs from 9 plant species. We found stable NLR root/shoot expression ratios within species, suggesting organ-specific hardwiring of NLR expression patterns in anticipation of distinct challenges. Most monocot and dicot plant species preferentially expressed NLRs in roots. In contrast, Brassicaceae species, including oilseed rape and the model plant Arabidopsis thaliana , were unique in showing NLR expression skewed towards the shoot across multiple phylogenetically distinct groups of NLRs. The Brassicaceae NLR expression shift coincides with loss of the endomycorrhization pathway, which enables intracellular root infection by symbionts. We propose that its loss offer two likely explanations for the unusual Brassicaceae NLR expression pattern: loss of NLR-guarded symbiotic components and elimination of constraints on general root defences associated with exempting symbionts from targeting. This hypothesis is consistent with the existence of Brassicaceae -specific receptors for conserved microbial molecules and suggests that Brassicaceae species are rich sources of unique antimicrobial root defences.

Abstract Correcting for sequencing and PCR errors is a major challenge when characterising genetic diversity using high-throughput amplicon sequencing (HTAS). Clustering amplicons by sequence similarity is a robust and frequently used approach, but it reduces sensitivity and makes it more difficult to detect differences between closely related strains. We have developed a multiplexed HTAS method, MAUI-seq, that incorporates unique molecular identifiers (UMIs) to improve correction. We profile Rhizobium leguminosarum biovar trifolii (Rlt) synthetic DNA mixes and Rlt diversity in white clover (Trifolium repens) root nodules by multiplexed sequencing of two plasmid-borne nodulation genes and two chromosomal genes. We show that the two main advantages of UMIs are efficient elimination of chimeric reads and the ability to confidently distinguish alleles that differ at only a single position. In addition, multi-amplicon profiling of genes from different replicons enables increased diversity profiling resolution, allowing us to demonstrate limited strain overlap between geographically distinct locations. The method does not rely on commercial library preparation kits and provides cost-effective, sensitive and flexible profiling of intra-species diversity.

Abstract Aim Lotus japonicus is a herbaceous perennial legume that has been used extensively as a genetically tractable model system for deciphering the molecular genetics of symbiotic nitrogen fixation. Our aim is to improve the L. japonicus reference genome sequence, which has so far been based on Sanger and Illumina sequencing reads from the L. japonicus accession MG-20 and contained a large fraction of unanchored contigs. Methods and Results Here, we use long PacBio reads from L. japonicus Gifu combined with Hi-C data and new high-density genetic maps to generate a high-quality chromosome-scale reference genome assembly for L. japonicus . The assembly comprises 554 megabases of which 549 were assigned to six pseudomolecules that appear complete with telomeric repeats at their extremes and large centromeric regions with low gene density. Conclusion and Perspectives The new L. japonicus Gifu reference genome and associated expression data represent valuable resources for legume functional and comparative genomics. Here, we provide a first example by showing that the symbiotic islands recently described in Medicago truncatula do not appear to be conserved in L. japonicus .

Faba bean (Vicia faba L.), a member of the Fabaceae family, is one of the important food legumes cultivated in cool temperate regions. It holds great importance for human consumption and livestock feed because of its high protein content, dietary fibre, and nutritional value. Major faba bean breeding challenges include its mixed breeding system, unknown wild progenitor, and genome size of ~13 Gb, which is the largest among diploid field crops. The key breeding objectives in faba bean include improved resistance to biotic and abiotic stress and enhanced seed quality traits. Major progress on reduction of vicine-convicine and seed coat tannins, the main anti-nutritional factors limiting faba bean seed usage, have been recently achieved through gene discovery. Genomic resources are relatively less advanced compared to other grain legume species, but significant improvements are underway due to a recent significant increase in research activities. A number of bi-parental populations have been constructed and mapped for targeted traits in the last decade. Faba bean now benefits from saturated synteny‐based genetic maps, along with next-generation sequencing and high-throughput genotyping technologies that are paving the way for marker-assisted selection. Developing a reference genome, and ultimately a pan-genome, will provide a foundational resource for molecular breeding. In this review, we cover the recent development and deployment of genomic tools for faba bean breeding.

Background: Blueberries are a rich source of antioxidants and other beneficial compounds that can protect against disease. Identifying genes involved in synthesis of bioactive compounds could enable breeding berry varieties with enhanced health benefits. Results: Toward this end, we annotated a draft blueberry genome assembly using RNA-Seq data from five stages of berry fruit development and ripening. Genome-guided assembly of RNA-Seq read alignments combined with output from ab initio gene finders produced around 60,000 gene models, of which more than half were similar to proteins from other species, typically the grape Vitis vinifera. Comparison of gene models to the PlantCyc database of metabolic pathway enzymes identified candidate genes involved in synthesis of bioactive compounds, including bixin, an apocarotenoid with potential disease-fighting properties, and defense-related cyanogenic glycosides, which are toxic. Cyanogenic glycoside (CG) biosynthetic enzymes were highly expressed in green fruit, and a candidate CG detoxification enzyme was up regulated during fruit ripening. Candidate genes for ethylene, anthocyanin, and 400 other biosynthetic pathways were also identified. Homology-based annotation using Blast2GO and InterPro assigned Gene Ontology terms to around 15,000 genes. RNA-Seq expression profiling showed that blueberry growth, maturation, and ripening involve dynamic gene expression changes, including coordinated up and down regulation of metabolic pathway enzymes and transcriptional regulators. Analysis of RNA-seq alignments identified developmentally regulated alternative splicing, promoter use, and 3? end formation. Conclusions: We report genome sequence, gene models, functional annotations, and RNA-Seq expression data that provide an important new resource enabling high throughput studies in blueberry. RNA-Seq data are freely available for visualization in Integrated Genome Browser, and analysis code is available from the git repository at http://bitbucket.org/lorainelab/blueberrygenome.

Andersen and Stougaard introduce the model legume Lotus japonicus.

The current Lotus japonicus reference genome sequence is based on a hybrid assembly of Sanger TAC/BAC, Sanger shotgun and Illumina shotgun sequencing data generated from the Miyakojima-MG20 accession. It covers nearly all expressed L. japonicus genes and has been annotated mainly based on transcriptional evidence. Analysis of repetitive sequences suggests that they are underrepresented in the reference assembly, reflecting an enrichment of gene-rich regions in the current assembly. Characterization of Lotus natural variation by resequencing of L. japonicus accessions and diploid Lotus species is currently ongoing, facilitated by the MG20 reference sequence.

Rhizobia supply legumes with fixed nitrogen using a set of symbiosis genes. These can cross rhizobium species boundaries, but it is unclear how many other genes show similar mobility. Here, we investigate inter-species introgression using de novo assembly of 196 Rhizobium leguminosarum bv. trifolii genomes. The 196 strains constituted a five-species complex, and we calculated introgression scores based on gene tree traversal to identify 171 genes that frequently cross species boundaries. Rather than relying on the gene order of a single reference strain, we clustered the introgressing genes into four blocks based on population structure-corrected linkage disequilibrium patterns. The two largest blocks comprised 125 genes and included the symbiosis genes, a smaller block contained 43 mainly chromosomal genes, and the last block consisted of three genes with variable genomic location. All introgression events were likely mediated by conjugation, but only the genes in the symbiosis linkage blocks displayed overrepresentation of distinct, high-frequency haplotypes. The three genes in the last block were core genes essential for symbiosis that had, in some cases, been mobilized on symbiosis plasmids. Inter-species introgression is thus not limited to symbiosis genes and plasmids, but other cases are infrequent and show distinct selection signatures.

Abstract Lotus japonicus is a herbaceous perennial legume that has been used extensively as a genetically tractable model system for deciphering the molecular genetics of symbiotic nitrogen fixation. So far, the L. japonicus reference genome assembly has been based on Sanger and Illumina sequencing reads from the L. japonicus accession MG-20 and contained a large fraction of unanchored contigs. Here, we use long PacBio reads from L. japonicus Gifu combined with Hi-C data and new high-density genetic maps to generate a high-quality chromosome-scale reference genome assembly for L. japonicus. The assembly comprises 554 megabases of which 549 were assigned to six pseudomolecules that appear complete with telomeric repeats at their extremes and large centromeric regions with low gene density. The new L. japonicus Gifu reference genome and associated expression data represent valuable resources for legume functional and comparative genomics. Here, we provide a first example by showing that the symbiotic islands recently described in Medicago truncatula do not appear to be conserved in L. japonicus.

Background: The influence of farming on plant, animal and microbial biodiversity has been carefully studied and much debated.Here, we compare an isolate-based study of 196 Rhizobium strains to amplicon-based MAUI-seq analysis of rhizobia from 17,000 white clover root nodules.We use these data to investigate the influence of soil properties, geographic distance, and field management on Rhizobium nodule populations.Results: Overall, there was good agreement between the two approaches and the precise allele frequency estimates from the large-scale MAUI-seq amplicon data allowed detailed comparisons of rhizobium populations between individual plots and fields.A few specific chromosomal core-gene alleles were significantly correlated with soil clay content, and core-gene allele profiles became increasingly distinct with geographic distance.Field management was associated with striking differences in Rhizobium diversity, where organic fields showed significantly higher diversity levels than conventionally managed trials.Conclusions: Our results indicate that MAUI-seq is suitable and robust for assessing nodule Rhizobium diversity.We further observe possible profound effects of field management on microbial diversity, which could impact plant health and productivity and warrant further investigation.

Abstract Legumes acquire access to atmospheric nitrogen through nitrogen fixation by rhizobia in root nodules. Rhizobia are soil dwelling organisms and there is a tremendous diversity of rhizobial species in different habitats. From the legume perspective, host range is a compromise between the ability to colonize new habitats, where the preferred symbiotic partner may be absent, and guarding against infection by suboptimal nitrogen fixers. Here, we investigate natural variation in rhizobial host range across Lotus species. We find that Lotus burttii is considerably more promiscuous than Lotus japonicus, represented by the Gifu accession, in its interactions with rhizobia. This promiscuity allows Lotus burttii to form nodules with Mesorhizobium, Rhizobium, Sinorhizobium, Bradyrhizobium , and Allorhizobium species that represent five distinct genera. Using recombinant inbred lines, we have mapped the Gifu/ burttii promiscuity QTL to the same genetic locus regardless of rhizobial genus, suggesting a general genetic mechanism for host-range expansion. The Gifu/ burttii QTL now provides an opportunity for genetic and mechanistic understanding of promiscuous legume-rhizobia interactions.

Since the genome sequence of Lotus japonicus, a model plant of family Fabaceae, was determined in 2008 (Sato et al. 2008), the genomes of other members of the Fabaceae family, soybean (Glycine max) (Schmutz et al. 2010) and Medicago truncatula (Young et al. 2011), have been sequenced. In this section, we introduce representative, publicly accessible online resources related to plant materials, integrated databases containing legume genome information, and databases for genome sequence and derived marker information of legume species including L. japonicus.

Background: Sequencing and PCR errors are a major challenge when characterising genetic diversity using high-throughput amplicon sequencing (HTAS).Results: We have developed a multiplexed HTAS method, MAUI-seq, which uses unique molecular identifiers (UMIs) to improve error correction by exploiting variation among sequences associated with a single UMI.We show that two main advantages of this approach are efficient elimination of chimeric and other erroneous reads, outperforming DADA2 and UNOISE3, and the ability to confidently recognise genuine alleles that are present at low abundance or resemble chimeras.Conclusions: The method provides sensitive and flexible profiling of diversity and is readily adaptable to most HTAS applications, including microbial 16S rRNA profiling and metabarcoding of environmental DNA.

Article Figures and data Abstract Editor's evaluation eLife digest Introduction Results Discussion Materials and methods Data availability References Decision letter Author response Article and author information Metrics Abstract Quinoa germplasm preserves useful and substantial genetic variation, yet it remains untapped due to a lack of implementation of modern breeding tools. We have integrated field and sequence data to characterize a large diversity panel of quinoa. Whole-genome sequencing of 310 accessions revealed 2.9 million polymorphic high confidence single nucleotide polymorphism (SNP) loci. Highland and Lowland quinoa were clustered into two main groups, with FST divergence of 0.36 and linkage disequilibrium (LD) decay of 6.5 and 49.8 kb, respectively. A genome-wide association study using multi-year phenotyping trials uncovered 600 SNPs stably associated with 17 traits. Two candidate genes are associated with thousand seed weight, and a resistance gene analog is associated with downy mildew resistance. We also identified pleiotropically acting loci for four agronomic traits important for adaptation. This work demonstrates the use of re-sequencing data of an orphan crop, which is partially domesticated to rapidly identify marker-trait association and provides the underpinning elements for genomics-enabled quinoa breeding. Editor's evaluation This is a comprehensive study of genomic and phenotypic diversity in the orphan crop quinoa. Based on whole genome resequencing of 310 accessions and field phenotyping of the same set of accessions for two years, the study identified the genetic basis of agronomically important traits. Based on this promising work, there will likely be scope for quick improvement of this orphan crop through breeding. https://doi.org/10.7554/eLife.66873.sa0 Decision letter Reviews on Sciety eLife's review process eLife digest As human populations grow and climate change tightens its grip, developing nutritious crops which can thrive on poor soil and under difficult conditions will become a priority. Quinoa, a harvest currently overlooked by agricultural research, could be an interesting candidate in this effort. With its high nutritional value and its ability to tolerate drought, frost and high concentrations of salt in the soil, this hardy crop has been cultivated in the Andes for the last 5,000 to 7,000 years. Today its commercial production is mainly limited to Peru, Bolivia, and Ecuador. Pinpointing the genetic regions that control traits such as yields or flowering time would help agronomists to create new varieties better suited to life under northern latitudes and mechanical farming. To identify these genes, Patiranage et al. grew 310 varieties of quinoa from all over the world under the same conditions; the genomes of these plants were also examined in great detail. Analyses were then performed to link specific genetic variations with traits relevant to agriculture, helping to pinpoint changes in the genetic code linked to differences in how the plants grew, resisted disease, or produced seeds of varying quality. Candidate genes likely to control these traits were then put forward. The study by Patiranage et al. provides a genetic map where genes of agronomical importance have been precisely located and their effects measured. This resource will help to select genetic profiles which could be used to create new quinoa breeds better adapted to a changing world. Introduction Climate change poses a great threat to crop production worldwide. In temperate climates of the world, higher temperatures and extended drought periods are expected. Moreover, crop production in industrialized countries depends on only a few major crops resulting in narrow crop rotations. Therefore, rapid transfer of wild species into crops using genetic modification and targeted mutagenesis is currently discussed (Li et al., 2018; Stetter et al., 2017). Alternatively, orphan crops with a long tradition of cultivation but low breeding intensity can be genetically improved by genomics-assisted selection methods. Quinoa (Chenopodium quinoa Willd.) is a pseudocereal crop species with a long history of cultivation. It was first domesticated about 5000–7000 years ago in the Andean region. Quinoa was a staple food during the pre-Columbian era, and the cultivation declined after the introduction of crops like wheat and barley by the Spanish rulers. Owing to diversity, biotic and abiotic stress tolerance, and ecological plasticity, quinoa can adapt to a broad range of agroecological regions (González et al., 2015; Ruiz et al., 2013). Due to its high seed protein content and favorable amino acid composition, the nutritional value is even higher than beef, fish, and other major cereals (Abugoch James, 2009; Vega-Gálvez et al., 2010). These favorable characteristics contributed to the increasing worldwide popularity of quinoa among consumers and farmers. A spontaneous hybridization event between two diploid species between 3.3 and 6.3 million years ago gave rise to the allotetraploid species quinoa (2n = 4x = 36) with a genome size of 1.45–1.5 Gb (nuclear DNA content 1C=1.49 pg) (Kolano et al., 2011; Palomino et al., 2008). A reference genome of the coastal Chilean quinoa accession PI 614886 has been published with 44,776 predicted gene models and whole-genome re-sequencing of Chenopodium pallidicaule and Chenopodium suecicum species, close relatives of the A and B subgenome donor species, respectively (Jarvis et al., 2017). The organellar genomes are originated from the A-genome ancestor (Maughan et al., 2019). Quinoa belongs to the Amaranthaceae, together with some other economically important crops like sugar beet, red beet, spinach, and amaranth. It reproduces sexually after self-pollination. Quinoa accessions are typically homozygous inbred lines. Nonetheless, heterozygosity in some accessions has been reported, indicating cross-pollination (Christensen et al., 2007). The inflorescences are panicles, which are often highly branched. Florets are tiny, which is a significant obstacle for hand-crossing. However, routine protocols for F1 seed production in combination with marker-assisted selection have been developed recently (Emrani et al., 2020; Peterson et al., 2015). Systematic breeding of quinoa is still in its infancy compared to major crops. Until recently, breeding has been mainly limited to Bolivia (Gandarillas, 1979) and Peru (Bonifacio et al., 2013), the major growing quinoa areas. Therefore, quinoa can be regarded as a partially domesticated crop. Many accessions suffer from seed shattering, branching, and non-appropriate plant height (PH), typical domestication traits. The major breeding objectives are apart from these characters: grain yield and seed size, downy mildew resistance, synchronized maturity, stalk strength, and low saponin content (Gomez‐Pando, 2015). In the past years, activities have been intensified to breed quinoa genotypes adapted to temperate environments, for example, Europe, North America, and China (Murphy, 2018). Here, the major problem is adapting to long-day conditions because quinoa is predominantly a short-day plant due to its origin from regions near the equator. There are only a few studies about the genetic diversity of quinoa. They were mainly based on phenotypic observations (Gandarillas et al., 1979; Wilson, 1988) and low throughput marker systems like random amplified polymorphic DNA (Ruas et al., 1999), amplification fragment length polymorphisms (Rodríguez and Isla, 2009), and microsatellites (Mason et al., 2005). Maughan et al., 2012, used five bi-parental populations to identify ca. 14,000 single nucleotide polymorphisms (SNPs), from which 511 KASP markers were developed. Genotyping 119 quinoa accessions gave the first insight into the population structure of this species (Maughan et al., 2012). Now, the availability of a reference genome enables genome-wide genotyping. Jarvis et al., 2017, re-sequenced 15 accessions and identified ca. 7.8 million SNPs. In another study, 11 quinoa accessions were re-sequenced, and 8 million SNPs and ca. 842,000 indels were identified (Zhang et al., 2017). Our study aimed to analyze the population structure of quinoa and patterns of variation by re-sequencing a diversity panel encompassing germplasm from all over the world. Using millions of markers, we performed a genome-wide association study using multiple year field data. Here, we identified QTLs (quantitative trait loci) that control agronomically important traits important for breeding cultivars to be grown under long-day conditions. Our results provide information for further understanding the genetic basis of agronomically important traits in quinoa and will be instrumental for future breeding. Results Re-sequencing 310 quinoa accessions reveal high sequence variation We assembled a diversity panel made of 310 quinoa accessions representing regions of major geographical distributions of quinoa (Figure 1—figure supplement 1). The diversity panel comprises accessions with different breeding histories (Supplementary file 1a). We included 14 accessions from a previous study, of which 7 are wild relatives (Jarvis et al., 2017). The mean mapped read depth ranged from 4.07 to 14.55, with an average of 7.78, indicating an adequate mapping quality required for accurate SNP calling despite the relatively modest sequencing depth. We mapped sequence reads to the reference genome V2 (CoGe id60716). Using mapping reads, we identified 45,330,710 unfiltered SNPs. After filtering the initial set of SNPs, we identified 4.5 million SNPs in total for the base SNP set. We further filtered the SNPs for MAF >5% (HCSNPs). We obtained 2.9 million high confidence SNPs for subsequent analysis (Supplementary file 1b). Across the whole genome, the average SNP density was 2.39 SNPs/kb. However, SNP densities were highly variable between genomic regions and ranged from 0 to 122 SNPs/kb (Figure 1—figure supplement 2). We did not observe significant differences in SNP density between the two subgenomes (a subgenome 2.43 SNPs/kb; B subgenome 2.35 SNPs/kb). Moreover, we did not see any correlation between sequencing depth and heterozygosity (Figure 1—figure supplement 2b), which indicates an adequate mapping quality required for accurate SNP calling. In an additional analysis, we divided the filtered SNPs into homozygous and heterozygous SNPs for each sample. Then, we calculated the mean read depth (DP) and genotype quality (GQ) of each sample separately for the homozygous and heterozygous fraction of the genome (Figure 1—figure supplement 3). Mean GQ of the heterozygous SNP calls was 61.34, whereas the mean GQ of homozygous SNP calls was 21.19, indicating that a higher stringency was used for the heterozygous SNP calls. We also compared the DP with the GQ for both filtered and unfiltered SNPs. The results indicated that higher GQ values were used for low DP regions in order to ensure correct genotype calls. Then, we split the SNPs by their functional effects as determined by SnpEff (Cingolani et al., 2012a). Among SNPs located in non-coding regions, 598,383 and 617,699 SNPs were located upstream (within 5 kb from the transcript start site) and downstream (within 5 kb from the stop site) of a gene, whereas 114,654 and 251,481 SNPs were located within exon and intron sequences, respectively (Table 1). We further searched for SNPs within coding regions. We found 70,604 missense SNPs and 41,914 synonymous SNPs within coding regions of 53,042 predicted gene models. Table 1 Summary statistics of genome-wide single nucleotide polymorphisms identified in 303 quinoa accessions. ParameterTypeAll genotypes(quinoa only)Highland populationLowland populationSNPTotal2,872,9352,590,9071,938,225Population-specific SNPs1,512,301859,619Intergenic2,452,3472,227,9521,649,310Introns251,481101,546172,692Exons114,654214,94578,248Nucleotide diversity5.78 × 10–43.56 × 10–4Tajima’s D0.884–0.384Population divergencesFST(weighted average)0.36 Linkage disequilibrium and population structure of the quinoa diversity panel Across the whole genome, linkage disequilibrium (LD) decay between SNPs averaged 32.4 kb (at r2=0.2). We did not observe substantial LD differences between subgenome A (r2=0.2 at 31.9 kb) and subgenome B (r2=0.2 at 30.7 kb) (Figure 1—figure supplement 4). The magnitude of LD decay among chromosomes did not vary drastically except for chromosome Cq6B, which exhibited a substantially slower LD decay (Figure 1—figure supplement 4a b). Then, we unraveled the population structure of the diversity panel. We performed principal component (PCA(SNP)), population structure, and phylogenetic analyses. PCA(SNP) showed two main clusters consistent with previous studies (Christensen et al., 2007). The first and second principal components (PC1(SNP) and PC2(SNP)) explained 23.35% and 9.45% of the variation, respectively (Figure 1a); 202 (66.67%) accessions were assigned to subpopulation 1 (SP1) and 101 (33.33%) to subpopulation 2 (SP2). SP1 comprised mostly Highland accessions, whereas Lowland accessions were found in SP2. PCA demonstrated a higher genetic diversity of the Highland population (Figure 1a). We also calculated PCs for each chromosome separately. For 16 chromosomes, the same clustering as for the whole genome was calculated. Nevertheless, two chromosomes, Cq6B and Cq8B, showed three distinct clusters (Figure 1—figure supplement 5). This is due to the split of the Lowland population into two clusters. We reasoned that gene introgressions on these two chromosomes from another interfertile group might have caused these differences. This is also supported by a slower LD decay on chromosome Cq6B (Figure 1—figure supplement 4b). This discrepancy also might arise due to the Lowland reference genome used for mapping the reads in this study (CoGe id60716), which may have structural differences compared to the genomes of Highland accessions. Figure 1 with 9 supplements see all Download asset Open asset Genetic diversity and population structure of the quinoa diversity panel. (a) Principal component analysis (PCA) of 303 quinoa accessions. PC1 and PC2 represent the first two analysis components, accounting for 23.35% and 9.45% of the total variation, respectively. The colors of dots represent the origin of accessions. Two populations are highlighted by different colors: Highland (light blue) and Lowland (pink). (b) Subpopulation-wise linkage disequilibrium (LD) decay in Highland (blue) and Lowland population (red). (c) Population structure is based on 10 subsets of SNPs, each containing 50,000 single nucleotide polymorphisms (SNPs) from the whole-genome SNP data. Model-based clustering was done in ADMIXTURE with different numbers of ancestral kinships (K=2 and K=8). K=8 was identified as the optimum number of populations. Left: Each vertical bar represents an accession, and color proportions on the bar correspond to the genetic ancestry. Right: Unrooted phylogenetic tree of the diversity panel. Colors correspond to the subpopulation. Figure 1—source data 1 Principal component analysis (PCA) of 303 quinoa accessions. https://cdn.elifesciences.org/articles/66873/elife-66873-fig1-data1-v2.xlsx Download elife-66873-fig1-data1-v2.xlsx Figure 1—source data 2 Subpopulation-wise linkage disequilibrium (LD) decay in Highland and Lowland population. https://cdn.elifesciences.org/articles/66873/elife-66873-fig1-data2-v2.xlsx Download elife-66873-fig1-data2-v2.xlsx Figure 1—source data 3 Population structure is based on 10 subsets of single nucleotide polymorphisms (SNPs), each containing 50,000 SNPs from the whole-genome SNP data. Model-based clustering was done in ADMIXTURE with different numbers of ancestral kinships. https://cdn.elifesciences.org/articles/66873/elife-66873-fig1-data3-v2.xlsx Download elife-66873-fig1-data3-v2.xlsx We also performed a population structure analysis with the ADMIXTURE software. We used 10 independent sets of 50,000 randomly chosen SNPs. Then, we performed ADMIXTURE analysis for each subset separately with a predefined number of genetic clusters K from 2 to 10 and different random seeds with 1000 bootstraps. The Q-matrices obtained were aligned with the greedy algorithm in the CLUMPP software (Jakobsson and Rosenberg, 2007). We used cross-validation to estimate the most suitable number of populations. Cross-validation error decreased as the K value increased, and we observed that after K=5, cross-validation error reached a plateau (Figure 1—figure supplement 6b). We observed allelic admixtures in some accessions, likely owing to their breeding history. The wild accessions were also clearly separated at the smallest cross-validation error of K=8, except two Chenopodium hircinum accessions (Figure 1c). This could be because C. hircinum is the closest crop wild relative; it also may have outcrossed with quinoa. The Highland population was structured into five groups, while the Lowland accessions were split into two subpopulations. The broad agro-climatic diversity of the Andean Highland germplasm might have caused a higher number of subpopulations. For clustering accessions based on sequence polymorphism, we combined 10 subsets created for ADMIXTURE analysis and removed redundant SNPs among subsets. We analyzed the relationships between quinoa accessions using 434,077 SNPs. Constructing a maximum likelihood (ML) tree gave rise to five clades (Figure 2). We found that the placement of the wild quinoa species as distant outgroups was concordant with the previous reports confirming that quinoa was domesticated from C. hircinum (Jarvis et al., 2017). However, we found that the C. hircinum accession BYU 566 (from Chile) was placed at the base of both Lowland and Highland clades, which contrasts to Jarvis et al., 2017, where this accession was placed at the base of Lowland (coastal) quinoa. As expected, accessions from the USA and Chile are closely related because the USDA germplasm had been collected at these geographical regions. Figure 2 with 1 supplement see all Download asset Open asset Maximum likelihood tree of 303 quinoa and 7 wild Chenopodium accessions from the diversity panel. Colors depict the geographical origin of accessions. Figure 2—source data 1 Maximum likelihood tree of 303 quinoa and 7 wild Chenopodium accessions from the diversity panel. https://cdn.elifesciences.org/articles/66873/elife-66873-fig2-data1-v2.xlsx Download elife-66873-fig2-data1-v2.xlsx Genomic patterns of variations between Highland and Lowland quinoa We were interested in patterns of variation in response to geographical diversification. We used PCA derived clusters and phylogenetic analysis to define two diverged quinoa populations (namely Highland and Lowland). These divergent groups are highly correlated with Highland and Lowland geographical origin. We used the base SNP set to analyze diversity statistics. To detect genomic regions affected by the population differentiation, we measured the level of nucleotide diversity using 10 kb non-overlapping windows (Varshney et al., 2017b; Figure 1—figure supplement 7). Then, we calculated the whole genome-wide LD decay across the two populations (Highland vs. Lowland); LD decayed more rapidly in Highland quinoa (6.5 kb vs. 49.8 kb, at r2=0.2) (Figure 1b). To measure nucleotide diversity, we scanned the quinoa genome with non-overlapping windows of 10 kb in length in both populations separately. The nucleotide diversity of the Highland population (5.78 × 10–4) was 1.62-fold higher compared to the Lowland population (3.56 × 10–4) (Table 1 and Figure 1—figure supplement 7). We observed left-skewed distribution and negative Tajima’s D value (–0.3883) in the Lowland populations indicating recent population growth (Table 1 and Figure 1—figure supplement 8). Genomic regions involved in adaptation to the Highlands should have much lower diversity in the Highland population than in the Lowland population, and genomic regions involved in adaptation to the Lowlands should have lower diversity in the Lowland population compared to the Highland population. Therefore, we calculated the nucleotide diversity ratios between Highland and Lowland to identify major genomic regions underlying the population differentiation (Figure 1—figure supplement 7). The FST value between populations was estimated to be 0.36, illustrating strong population differentiation. Concerning the regions of variants, exonic SNPs are substantially higher in the Highland population (Table 1 and Figure 1—figure supplement 7). We conducted a local PCA to identify genomic regions with a strong population structure. The genome was divided into 50 kb non-overlapping windows, and PCA was calculated for each window using the lostruct program (Li and Ralph, 2019), which calculates a similarity score by comparing PCs obtained from each window. Similarity scores were then stored as a matrix and visualized using multidimensional scaling (MDS) transformation. Strong indications of the population structure are represented in the corners of the MDS analysis; usually, it follows a triangle, providing three corners (corner 1, corner 2, and corner 3) (Figure 1—figure supplement 9a). Candidate genomic regions were defined as the 1% of the MDS coordinates closest to each of the corners. They consist of the windows with the strongest genetic differentiation across the genome (Figure 1—figure supplement 9b). Then, we selected candidate genomic regions from each corner and calculated the PCs using SNPs present in those regions. SNPs from the candidate genomic regions of corner 1 structured the diversity panel into two clusters (Figure 1—figure supplement 9c). Corner 2 also resulted in two clusters, but clustering was not as strong as corner 1 regions (Figure 1—figure supplement 9d). Corner 3 separated accessions into three clusters similar to the PCA using the candidate genomic regions obtained from the nucleotide diversity ratio analysis (Figure 1—figure supplement 9f and g). Then, we located the genes within candidate regions obtained from all three analyses. We identified 936, 953, and 546 candidate genes located within critical regions from the nucleotide diversity ratio (π (Highland/Lowland)), FST, and local PCA corner 1 (Figure 1—figure supplement 9h). Of these, only four genes were shared among all analyses, and 30 genes were shared between FST and genomic regions in corner 1 of the local PCA plot. Genomic regions in corner 3 of the local PCA plot and with a high nucleotide diversity ratio shared 102 genes (Figure 1—figure supplement 9i). Mapping agronomically important trait loci in the quinoa genome We evaluated 13 qualitative and 4 dichotomous traits on 350 accessions across 2 different years. At the time of the final harvest, 254 accessions did not yet reach maturity (senescence). However, all accessions produced seeds, therefore they could be investigated for seed-related traits. For all traits, substantial phenotypic variation among accessions was found. High heritabilities were calculated for all quantitative traits except for the number of branches (NoB) and stem lying (STL), which indicates that the phenotypic variation between the accessions is caused mainly by genetic variation (Supplementary file 1c). Trait correlations between years were also high (Figure 3—figure supplement 2), which is following the heritability estimates. We found the strongest positive correlation between days to maturity (DTM) and panicle length (PL), and PH, whereas the strongest negative correlation was found between DTM and thousand seed weight (TSW) (Figure 3—figure supplement 3). Then, a PCA was performed based on 12 quantitative traits (PCA(PHEN)) to explore the phenotypic relationship among quinoa accessions. The first two PCs explained 62.12% of the phenotypic variation between the accessions. The score plot of the PCs showed a similar clustering pattern as the SNP-based PCA (PCA(SNP)) (Figure 1 and Figure 3—figure supplement 4a). PCA(PHEN) variables factor map indicated that most Lowland accessions were high yielding with high TSW and dense panicles. Moreover, these accessions are early flowering and early maturing, and they are short (Figure 3—figure supplement 4b). Phenotype-based PCA(PHEN) also showed that the Lowland accessions are better adapted/selected for cultivation in long-day photoperiods than the Highland accessions. These results are in accordance with LD, nucleotide diversity, and Tajima’s D estimations, implying the Lowland accessions underwent a stronger selection during breeding. Then, we calculated the best linear unbiased estimates (BLUE) of the traits investigated. In total, 294 accessions shared the re-sequencing information and phenotypes out of 350 phenotypically evaluated accessions. For GWAS analysis, we used ~2.9 million high confidence SNPs. We considered pairwise kinship value distribution to determine that all accessions could be used for GWAS analysis without conducting subpopulation-wise analyses (Figure 3—source data 4). In total, we identified 1480 significant (suggestive threshold: p<9.41e-7) SNP-trait associations (marker-trait association [MTAs]) for 17 traits (Figure 3—source data 4). The number of MTAs ranged from 4 (STL) to 674 (DTM) (Supplementary file 1d). In agreement with previous reports, we defined an MTA as ‘consistent’ when detected in both years (Varshney et al., 2019). We identified 600 consistent MTAs across 11 traits. TSW and DTM showed the highest number of ‘consistent’ associations. Among these, 143 MTAs are located within a gene, and 22 SNPs resulted in a missense mutation (Supplementary file 1e). MTAs for the duration from bolting to flowering (days to bolting to days to flowering), number of branching, seed yield, STL, and growth type (GT) were not ‘consistent’ between years (Figure 3—source data 4). This is also reflected by the low estimates of heritability for these traits, indicating considerably higher genotype × environment interactions. Using the SNPs not located in the repetitive regions, we identified 619 MTA across 11 traits, of which 291 MTAs are common between both analyses (Figure 3—figure supplement 7). Unique associations of 476 and 328 were identified from whole-genome SNPs and repeat masked SNPs, respectively. However, the comparison of GWAS results of PCs between whole-genome SNP set and repeat masked SNP set showed that highly significant associations could be identified even if the repetitive regions were excluded from the analysis (Figure 3—figure supplement 7b and c). Candidate genes for agronomically important traits First, we tested the resolution of our mapping study. We searched for candidate genes 50 kb down- and upstream of significant SNPs for two qualitative traits in quinoa, flower color and seed saponin content. To define candidate genes, we considered homologous genes that have already been functionally characterized in other species. We identified highly significant MTAs for stem color on chromosome Cq1B (69.72–69.76 Mb). Two genes (CqCYP76AD1 and CqDODA1) in this region exhibit high sequence homology to betalain synthesis pathway genes BvCYP76AD1 (Hatlestad et al., 2012) and BvDODA1 (Bean et al., 2018) from sugar beet (Figure 4—figure supplement 1). A significant MTA for saponin content on chromosome Cq5B between 8.85 and 9.2 Mb included 29 genes, of which the two BHLH25 were in LD with the significantly associated SNPs. BHLH25 genes were reported to control saponin content in quinoa (Jarvis et al., 2017; Figure 4—figure supplement 1b). This demonstrates that the marker density is high enough to narrow down to causative genes underlying a trait. Then, we examined four quantitative traits. We obtained seven MTAs on chromosome Cq2A with the traits days to flowering, DTM, PH, and PL indicating pleiotropic gene action (Figure 3a and Supplementary file 1f). To further investigate genes that are pleiotropically active on different traits, we followed a multivariate approach (Solovieff et al., 2013). First, we performed a PCA of the four phenotypes (cross-phenotypes; genetically correlated traits). We found 89.94% of the variation could be explained by the first two PCs of the cross-phenotypes (PCA(CP)) (Figure 3—figure supplement 5), which suggests that PCA (CP) is suitable to reduce dimensions for a GWAS of cross-phenotypes. Since the PCA(CP) revealed a similar clustering as PCA(SNP), these analyses results provide preliminary indications that in quinoa, days to flowering, DTM, PH, and PL are highly associated with population structure and may reflect adaptation to diverse environments. Then, we performed a GWAS analysis using the first three PCs as traits (PC-GWAS) (Figure 3—figure supplement 5c). We identified strong associations on chromosomes Cq2A, Cq7B (PC1), and Cq8B (PC2) (Figure 3—figure supplement 6). Of 468 MTAs (PC1: 426 and PC2: 42) across the whole genome, 222 (PC1: 211 and PC2: 11) are located within 95 annotated genes. We found 14 SNPs that changed the amino acid sequence in 12 predicted protein sequences of associated genes (Supplementary file 1e). In the next step, we searched genes located 50 kb flanking to an MTA, considering a threshold that is below the genome-wide LD of the Lowland population. Altogether, 605 genes were identified (PC1: 520 and PC2: 85) (Supplementary file 1g). Figure 3 with 7 supplements see all Download asset Open asset Genomic regions associated with important agronomic traits. (a) Significant marker-trait associations (MTAs) for days to flowering (DTF), days to maturity (DTM), plant height (PH), and panicle density on chromosome Cq2A. Red color arrows indicate the single nucleotide polymorphism (SNP) loci pleiotrop