The sequence variation present in accessions conserved in genebanks can best be used in plant improvement when it is properly characterized and published. Using low cost and high density single nucleotide polymorphism (SNP) assays, the genetic diversity, population structure, and relatedness between pairs of accessions can be quickly assessed. This information is relevant for different purposes, including creating core and mini-core sets that represent the maximum possible genetic variation contained in the whole collection. Here, we studied the genetic variation and population structure of 2,179 Oryza glaberrima Steud. accessions conserved at the AfricaRice genebank using 27,560 DArTseq-based SNPs. Only 14% (3,834 of 27,560) of the SNPs were polymorphic across the 2,179 accessions, which is much lower than diversity reported in other Oryza species. Genetic distance between pairs of accessions varied from 0.005 to 0.306, with 1.5% of the pairs nearly identical, 8.0% of the pairs similar, 78.1% of the pairs moderately distant, and 12.4% of the pairs very distant. The number of redundant accessions that contribute little or no new genetic variation to the O. glaberrima collection was very low. Using the maximum length sub-tree method, we propose a subset of 1,330 and 350 accessions to represent a core and mini-core collection, respectively. The core and mini-core sets accounted for ~61 and 16%, respectively, of the whole collection, and captured 97–99% of the SNP polymorphism and nearly all allele and genotype frequencies observed in the whole O. glaberrima collection available at the AfricaRice genebank. Cluster, principal component and model-based population structure analyses all divided the 2,179 accessions into five groups, based roughly on country of origin but less so on ecology. The first, third and fourth groups consisted of accessions primarily from Liberia, Nigeria, and Mali, respectively; the second group consisted primarily of accessions from Togo and Nigeria; and the fifth and smallest group was a mixture of accessions from multiple countries. Analysis of molecular variance showed between 10.8 and 28.9% of the variation among groups with the remaining 71.1–89.2% attributable to differences within groups.

Species misclassification (misidentification) and handling errors have been frequently reported in various plant species conserved at diverse gene banks, which could restrict use of germplasm for correct purpose. The objectives of the present study were to (i) determine the extent of genotyping error (reproducibility) on DArTseq-based single-nucleotide polymorphisms (SNPs); (ii) determine the proportion of misclassified accessions across 3134 samples representing three African rice species complex (Oryza glaberrima, O. barthii, and O. longistaminata) and an Asian rice (O. sativa), which are conserved at the AfricaRice gene bank; and (iii) develop species- and sub-species (ecotype)-specific diagnostic SNP markers for rapid and low-cost quality control (QC) analysis. Genotyping error estimated from 15 accessions, each replicated from 2 to 16 times, varied from 0.2 to 3.1%, with an overall average of 0.8%. Using a total of 3134 accessions genotyped with 31,739 SNPs, the proportion of misclassified samples was 3.1% (97 of the 3134 accessions). Excluding the 97 misclassified accessions, we identified a total of 332 diagnostic SNPs that clearly discriminated the three indigenous African species complex from Asian rice (156 SNPs), O. longistaminata accessions from both O. barthii and O. glaberrima (131 SNPs), and O. sativa spp. indica from O. sativa spp. japonica (45 SNPs). Using chromosomal position, minor allele frequency, and polymorphic information content as selection criteria, we recommended a subset of 24 to 36 of the 332 diagnostic SNPs for routine QC genotyping, which would be highly useful in determining the genetic identity of each species and correct human errors during routine gene bank operations.Electronic supplementary materialThe online version of this article (10.1007/s11032-018-0885-z) contains supplementary material, which is available to authorized users.

Using interspecific crosses involving Oryza glaberrima Steud. as donor and O. sativa L. as recurrent parents, rice breeders at the Africa Rice Center developed several ‘New Rice for Africa (NERICA)’ improved varieties. A smaller number of interspecific and intraspecific varieties have also been released as ‘Advanced Rice for Africa (ARICA)’. The objective of the present study was to investigate the genetic variation, relatedness, and population structure of 330 widely used rice genotypes in Africa using DArTseq-based single nucleotide polymorphisms (SNPs). A sample of 11 ARICAs, 85 NERICAs, 62 O. sativa spp. japonica, and 172 O. sativa spp. indica genotypes were genotyped with 27,560 SNPs using diversity array technology (DArT)-based sequencing (DArTseq) platform. Nearly 66% of the SNPs were polymorphic, of which 15,020 SNPs were mapped to the 12 rice chromosomes. Genetic distance between pairs of genotypes that belong to indica, japonica, ARICA, and NERICA varied from 0.016 to 0.623, from 0.020 to 0.692, from 0.075 to 0.763, and from 0.014 to 0.644, respectively. The proportion of pairs of genotypes with genetic distance > 0.400 was the largest within NERICAs (35.1% of the pairs) followed by ARICAs (18.2%), japonica (17.4%), and indica (5.6%). We found one pair of japonica, 11 pairs of indica, and 35 pairs of NERICA genotypes differing by <2% of the total scored alleles, which was due to 26 pairs of genotypes with identical pedigrees. Cluster analysis, principal component analysis, and the model-based population structure analysis all revealed two distinct groups corresponding to the lowland (primarily indica and lowland NERICAs) and upland (japonica and upland NERICAs) growing ecologies. Most of the interspecific lowland NERICAs formed a sub-group, likely caused by differences in the O. glaberrima genome as compared with the indica genotypes. Analysis of molecular variance revealed very great genetic differentiation (FST = 0.688) between the lowland and upland ecologies, and 31.2% of variation attributable to differences within cluster groups. About 8% (1,197 of 15,020) of the 15,020 SNPs were significantly (P < 0.05) different between the lowland and upland ecologies and formed contrasting haplotypes that could clearly discriminate lowland from upland genotypes. This is the first study using high density markers that characterized NERICA and ARICA varieties in comparison with indica and japonica varieties widely used in Africa, which could aid rice breeders on parent selection for developing new improved rice germplasm.

to minimize the cost of sample preparation and genotyping, most genebank genomics studies in self-pollinating species are conducted on a single individual to represent an accession, which may be heterogeneous with larger than expected intra-accession genetic variation. Here, we compared various population genetics parameters among six DNA (leaf) sampling methods on 90 accessions representing a wild species (O. barthii), cultivated and landraces (O. glaberrima, O. sativa), and improved varieties derived through interspecific hybridizations. A total of 1,527 DNA samples were genotyped with 46,818 polymorphic single nucleotide polymorphisms (SNPs) using DArTseq. Various statistical analyses were performed on eleven datasets corresponding to 5 plants per accession individually and in a bulk (two sets), 10 plants individually and in a bulk (two sets), all 15 plants individually (one set), and a randomly sampled individual repeated six times (six sets). Overall, we arrived at broadly similar conclusions across 11 datasets in terms of SNP polymorphism, heterozygosity/heterogeneity, diversity indices, concordance among genetic dissimilarity matrices, population structure, and genetic differentiation; there were, however, a few discrepancies between some pairs of datasets. Detailed results of each sampling method, the concordance in their outputs, and the technical and cost implications of each method were discussed.

Morphological identification of closely related rice species, particularly those in the Oryza AA genome group, presents major challenges and often results in cases of misidentification. Recent work by this group identified diagnostic single nucleotide polymorphic (SNP) markers specific for several rice species and subspecies based on DArTseq next-generation sequencing technology (“DArTseq”). These SNPs can be used for quality control (QC) analysis in rice breeding and germplasm maintenance programs. Here, we present the DArTseq-based diagnostic SNPs converted into Kompetitive allele-specific PCR (KASPar or KASP) assays and validation data for a subset of them; these can be used for low-cost routine genotyping quality control (QC) analysis. Of the 224 species/subspecies’ diagnostic SNPs tested, 158 of them produced working KASP assays, a conversion success rate of 70%. Two validation experiments were run with 87 of the 158 SNP markers to ensure that the assays amplified, were polymorphic, and distinguished the five species/subspecies tested. Based on these validation test results, we recommend a panel of 36 SNP markers that clearly delineate O. barthii, O. glaberrima, O. longistaminata, O. sativa spp. indica and japonica. The KASP assays provide a flexible, rapid turnaround and cost-effective tool to facilitate germplasm curation and management of these four Oryza AA genome species across multiple genebanks.