Classification of Grain Amaranths Using Chromosome-Level Genome Assembly of Ramdana, A. hypochondriacus

North American pitseed goosefoot (Chenopodium berlandieri) is a genetic resource to improve Andean quinoa (C. quinoa)

Pitseed goosefoot (Chenopodium berlandieri) is a free-living North American member of an allotetraploid complex that includes the Andean pseudocereal quinoa (C. quinoa). Like quinoa, pitseed goosefoot was domesticated, possibly independently, in eastern North America (subsp. jonesianum) and Mesoamerica (subsp. nuttaliae). To test the utility of C. berlandieri as a resource for quinoa breeding, we produced the whole-genome DNA sequence of PI 433,231, a huauzontle from Puebla, México. The 1.295 Gb genome was assembled into 18 pseudomolecules and annotated using RNAseq data from multiple tissues. Alignment with the v.2.0 genome of Chilean-origin C. quinoa cv. 'QQ74' revealed several inversions and a 4A-6B reciprocal translocation. Despite these rearrangements, some quinoa x pitseed goosefoot crosses produce highly fertile hybrids with faithful recombination, as evidenced by a high-density SNP linkage map constructed from a Bolivian quinoa 'Real-1' × BYU 937 (Texas coastal pitseed goosefoot) F2 population. Recombination in that cross was comparable to a 'Real-1' × BYU 1101 (Argentine C. hircinum) F2 population. Furthermore, SNP-based phylogenetic and population structure analyses of 90 accessions supported the hypothesis of multiple independent domestications and descent from a common 4 × ancestor, with a likely North American Center of Origin.

Differentiation of Andean and Mesoamerican accessions in a proposed core collection of grain amaranths

Grain amaranths are made up of three New World species of pseudo-cereals with C4 photosynthesis from the dicotyledonous family Amaranthaceae and the genus Amaranthus. They originate in two ecoregions of the Americas, namely, the inter-Andean valleys of South America and the volcanic axis and lowlands of Mexico and Central America. These correspond to two centers of domestications for Andean and Mesoamerican crops, with one cultivated species found in the first region and two found in the latter region. To date, no core collection has been made for the grain amaranths in the United States Department of Agriculture (USDA) germplasm system. In this study, our objective was to create a core for the 2,899 gene bank accessions with collection site data by town or farm site of which 1,090 have current geo-referencing of latitude and longitude coordinates. We constituted the core with 260 genotypes of Amaranthus, which we evaluated with 90 single-nucleotide polymorphism markers. Our goal was to distinguish between Andean and Mesoamerican gene pools of amaranths, including the cultivated species and three possible progenitor or wild relative ancestors along with two more species in an outgroup. Population structure, clustering, and discriminant analysis for principal components showed that Andean species Amaranthus caudatus and Amaranthus quitensis shared fewer alleles with Mesoamerican species Amaranthus cruentus and Amaranthus hypochondriacus, compared to each group individually. Amaranthus hybridus was a bridge species that shared alleles with both regions. Molecular markers have the advantage over morphological traits at quickly distinguishing the Andean and Mesoamerican cultivars and have the added benefit of being useful for following inter-species crosses and introgression.

Amaranthus hypochondriacus seeds as a rich source of cysteine rich bioactive peptides

Amaranthus hypochondriacus is a nutritious alternative grain native to Central and South America. Increased interest in the impact of A. hypochondriacus on the human body has driven characterization of bioactive secondary metabolites. The seeds are known to contain bioactive small molecules but little is known regarding endogenous peptides. Cysteine-rich peptides (CRPs) in foodstuffs are particularly relevant because they are stabilized by disulfide bonds enhancing resistance to digestion. Here, in silico predictions, proteomics, and simulated gastrointestinal digestions are leveraged to identify digestion resistant CRPs within A. hypochondriacus seeds. Thirteen in silico predicted CRPs were detected in a seed extract providing evidence for the translation of five CRP families. Mature forms of six CRPs were characterized via top-down proteomics revealing multiple post-translational modifications. All six peptides demonstrated resistance to simulated gastrointestinal digestion, suggesting that A. hypochondriacus CRPs may exhibit bioactivity after consumption and should be prioritized for further characterization.

PPA-GCN: A Efficient GCN Framework for Prokaryotic Pathways Assignment

With the rapid development of sequencing technology, completed genomes of microbes have explosively emerged. For a newly sequenced prokaryotic genome, gene functional annotation and metabolism pathway assignment are important foundations for all subsequent research work. However, the assignment rate for gene metabolism pathways is lower than 48% on the whole. It is even lower for newly sequenced prokaryotic genomes, which has become a bottleneck for subsequent research. Thus, the development of a high-precision metabolic pathway assignment framework is urgently needed. Here, we developed PPA-GCN, a prokaryotic pathways assignment framework based on graph convolutional network, to assist functional pathway assignments using KEGG information and genomic characteristics. In the framework, genomic gene synteny information was used to construct a network, and ideas of self-supervised learning were inspired to enhance the framework’s learning ability. Our framework is applicable to the genera of microbe with sufficient whole genome sequences. To evaluate the assignment rate, genomes from three different genera (Flavobacterium (65 genomes) and Pseudomonas (100 genomes), Staphylococcus (500 genomes)) were used. The initial functional pathway assignment rate of the three test genera were 27.7% (Flavobacterium), 49.5% (Pseudomonas) and 30.1% (Staphylococcus). PPA-GCN achieved excellence performance of 84.8% (Flavobacterium), 77.0% (Pseudomonas) and 71.0% (Staphylococcus) for assignment rate. At the same time, PPA-GCN was proved to have strong fault tolerance. The framework provides novel insights into assignment for metabolism pathways and is likely to inform future deep learning applications for interpreting functional annotations and extends to all prokaryotic genera with sufficient genomes.