Vicia faba L, is a globally important grain legume whose main centers of diversity are the Fertile Crescent and Mediterranean basin. Because of its small number (six) of exceptionally large and easily observed chromosomes it became a model species for plant cytogenetics the 70s and 80s. It is somewhat ironic therefore, that the emergence of more genomically tractable model plant species such as Arabidopsis and Medicago coincided with a marked decline in genome research on the formerly favored plant cytogenetic model. Thus, as ever higher density molecular marker coverage and dense genetic and even complete genome sequence maps of key crop and model species emerged through the 1990s and early 2000s, genetic and genome knowledge of Vicia faba lagged far behind other grain legumes such as soybean, common bean and pea. However, cheap sequencing technologies have stimulated the production of deep transcriptome coverage from several tissue types and numerous distinct cultivars in recent years. This has permitted the reconstruction of the faba bean meta-transcriptome and has fueled development of extensive sets of Simple Sequence Repeat and Single Nucleotide Polymorphism (SNP) markers. Genetics of faba bean stretches back to the 1930s, but it was not until 1993 that DNA markers were used to construct genetic maps. A series of Random Amplified Polymorphic DNA-based genetic studies mainly targeted at quantitative loci underlying resistance to a series of biotic and abiotic stresses were conducted during the 1990's and early 2000s. More recently, SNP-based genetic maps have permitted chromosome intervals of interest to be aligned to collinear segments of sequenced legume genomes such as the model legume Medicago truncatula, which in turn opens up the possibility for hypotheses on gene content, order and function to be translated from model to crop. Some examples of where knowledge of gene content and function have already been productively exploited are discussed. The bottleneck in associating genes and their functions has therefore moved from locating gene candidates to validating their function and the last part of this review covers mutagenesis and genetic transformation, two complementary routes to validating gene function and unlocking novel trait variation for the improvement of this important grain legume.

Increasing the proportion of locally produced plant protein in currently meat-rich diets could substantially reduce greenhouse gas emissions and loss of biodiversity1. However, plant protein production is hampered by the lack of a cool-season legume equivalent to soybean in agronomic value2. Faba bean (Vicia faba L.) has a high yield potential and is well suited for cultivation in temperate regions, but genomic resources are scarce. Here, we report a high-quality chromosome-scale assembly of the faba bean genome and show that it has expanded to a massive 13 Gb in size through an imbalance between the rates of amplification and elimination of retrotransposons and satellite repeats. Genes and recombination events are evenly dispersed across chromosomes and the gene space is remarkably compact considering the genome size, although with substantial copy number variation driven by tandem duplication. Demonstrating practical application of the genome sequence, we develop a targeted genotyping assay and use high-resolution genome-wide association analysis to dissect the genetic basis of seed size and hilum colour. The resources presented constitute a genomics-based breeding platform for faba bean, enabling breeders and geneticists to accelerate the improvement of sustainable protein production across the Mediterranean, subtropical and northern temperate agroecological zones.

Faba bean ( Vicia faba L.) is a widely adapted and high-yielding legume cultivated for its protein-rich seeds ( 1 ). However, the seeds accumulate the pyrimidine glucosides vicine and convicine, which can cause hemolytic anemia (favism) in 400 million genetically predisposed individuals ( 2 ). Here, we use gene-to-metabolite correlations, gene mapping, and genetic complementation to identify VC1 as a key enzyme in vicine and convicine biosynthesis. We demonstrate that VC1 has GTP cyclohydrolase II activity and that the purine GTP is a precursor of both vicine and convicine. Finally, we show that cultivars with low vicine and convicine levels carry an inactivating insertion in the coding sequence of VC1 . Our results reveal an unexpected, purine rather than pyrimidine, biosynthetic origin for vicine and convicine and pave the way for the development of faba bean cultivars that are free of these anti-nutrients.

25Faba bean is a widely adapted and high-yielding legume cultivated for its protein-rich seeds 1 . 26 However, the seeds accumulate the anti-nutritional pyrimidine glucosides vicine and convicine, which 27 can cause haemolytic anaemia-favism-in the 400 million individuals genetically predisposed by a 28 deficiency in glucose-6-phosphate dehydrogenase 2 . Here, we identify the first enzyme associated with 29 vicine and convicine biosynthesis, which we name VC1. We show that VC1 co-locates with the major QTL 30 for vicine and convicine content and that the expression of VC1 correlates highly with vicine content 31 across tissues. We also show that low-vicine varieties express a version of VC1 carrying a small, frame-32 shift insertion, and that overexpression of wild-type VC1 leads to an increase in vicine levels. VC1 33 encodes a functional GTP cyclohydrolase II, an enzyme normally involved in riboflavin biosynthesis from 34 the purine GTP. Through feeding studies, we demonstrate that GTP is a precursor of vicine both in faba 35 bean and in the distantly related plant bitter gourd. Our results reveal an unexpected biosynthetic origin 36 for vicine and convicine and pave the way for the development of faba bean cultivars that are free from 37 these anti-nutrients, providing a safe and sustainable source of dietary protein. 38 39 Main Text 40 According to the UN's Intergovernmental Panel on Climate Change (IPCC), switching to a plant-based 41 diet can reduce carbon emissions, especially in the West 3 . The suggested change in diet will require a wider 42and more varied cultivation of locally adapted protein crops. On a worldwide basis, faba bean ( Fig. 1a) has 43 the highest yield of the legumes after soybean (1.92 Mg/ha in 2013-2017) 4 and the highest seed protein 44 content of the starch-containing legumes (29% dry-matter basis) 5 . Furthermore, faba bean is adapted to 45 cool climates such as Mediterranean winters and northern European summers, where soybean performs 46 poorly 6 . The main factor restricting faba bean cultivation and consumption is the presence of the anti-47 2, Supplementary File 6). Using seed coat cDNA and PCR primers able to distinguish between VC1 and vc1, 132

Key message We identified marker-trait associations for key faba bean agronomic traits and genomic signatures of selection within a global germplasm collection. Abstract Faba bean (Vicia faba L.) is a high-protein grain legume crop with great potential for sustainable protein production. However, little is known about the genetics underlying trait diversity. In this study, we used 21,345 high-quality SNP markers to genetically characterize 2678 faba bean genotypes. We performed genome-wide association studies of key agronomic traits using a seven-parent-MAGIC population and detected 238 significant marker-trait associations linked to 12 traits of agronomic importance. Sixty-five of these were stable across multiple environments. Using a non-redundant diversity panel of 685 accessions from 52 countries, we identified three subpopulations differentiated by geographical origin and 33 genomic regions subjected to strong diversifying selection between subpopulations. We found that SNP markers associated with the differentiation of northern and southern accessions explained a significant proportion of agronomic trait variance in the seven-parent-MAGIC population, suggesting that some of these traits were targets of selection during breeding. Our findings point to genomic regions associated with important agronomic traits and selection, facilitating faba bean genomics-based breeding.

Increasing the proportion of locally produced plant protein in currently meat-rich diets could substantially reduce greenhouse gas emission and loss of biodiversity. However, plant protein production is hampered by the lack of a cool-season legume equivalent to soybean in agronomic value. Faba bean (Vicia faba L.) has a high yield potential and is well-suited for cultivation in temperate regions, but genomic resources are scarce. Here, we report a high-quality chromosome-scale assembly of the faba bean genome and show that it has grown to a massive 13 Gb in size through an imbalance between the rates of amplification and elimination of retrotransposons and satellite repeats. Genes and recombination events are evenly dispersed across chromosomes and the gene space is remarkably compact considering the genome size, though with significant copy number variation driven by tandem duplication. Demonstrating practical application of the genome sequence, we develop a targeted genotyping assay and use high-resolution genome-wide association (GWA) analysis to dissect the genetic basis of hilum colour. The resources presented constitute a genomics-based breeding platform for faba bean, enabling breeders and geneticists to accelerate improvement of sustainable protein production across Mediterranean, subtropical, and northern temperate agro-ecological zones.

Faba bean (Vicia faba L.) is a high-protein grain legume crop with great potential for further cultivation. However, little is known about the genetics underlying trait diversity. In this study, we use 21,345 high-quality SNP markers to genetically characterise 2,678 faba bean genotypes. We perform genome-wide association studies of key agronomic traits using a Seven-parent-MAGIC population and detect 238 significant marker-trait associations linked to 12 traits of agronomic importance, with 65 of these being stable across multiple environments. Using a non-redundant diversity panel of 685 accessions from 52 countries, we identify 3 subpopulations differentiated by geographical origin and 33 genomic regions subject to strong diversifying selection between subpopulations. We find that SNP markers associated with the differentiation of northern and southern accessions were able to explain a significant proportion of agronomic trait variance in the Seven-parent-MAGIC population, suggesting that some of these traits have played an important role in breeding. Altogether, our findings point to genomic regions associated with important agronomic traits and selection in faba bean, which can be used for breeding purposes.Key MessageWe identified marker-trait associations for key faba bean agronomic traits and genomic signatures of selection within a global germplasm collection.

No abstract