Genetic Diversity, Population Structure and Subset Development in a Sesbania sesban Collection.

Sesbania sesban (L.) Merr. is a multipurpose legume tree grown primarily for fodder and forage in the tropical and subtropical world. In this study, the Sesbania sesban collection maintained in the International Livestock Research Institute (ILRI) forage Genebank was studied using genome-wide markers generated on the DArTseq platform. Genotyping produced 84,673 and 60,626 SNP and SilicoDArT markers with a mean polymorphic information content of 0.153 and 0.123, respectively. From the generated markers, 7587 and 15,031 highly informative SNP and SilicoDArT markers, respectively, were filtered and used for genetic diversity analysis and subset development. Analysis of molecular variance (AMOVA) revealed higher variability 'within' (52.73% for SNP markers and 67.36% for SilicoDArT markers) than 'between' accessions. Hierarchical cluster analysis showed the presence of four main clusters in the collection. Mantel correlation analysis showed a lack of relationship between genetic variation of the germplasm and their geographical origin. A representative subset of 34 accessions containing germplasm from diverse origins and agro-ecologies was developed using SNP markers. The genetic diversity information generated in this study could be used for marker-assisted screening for stress tolerance, gap analysis and identification and acquisition of new distinct genotype(s) to broaden the genetic basis of the collection for future improvement programs to develop high-yielding, stress-tolerant varieties for enhancing food and environmental security in crop-livestock-based production systems.

Genetic Diversity and Population Structure of a Rhodes Grass (Chloris gayana) Collection.

Rhodes grass (Chloris gayana Kunth) is one of the most important forage grasses used throughout the tropical and subtropical regions of the world. Enhancing the conservation and use of genetic resources requires the development of knowledge and understanding about the existing global diversity of the species. In this study, 104 Rhodes grass accessions, held in trust in the ILRI forage genebank, were characterized using DArTSeq markers to evaluate the genetic diversity and population structure, and to develop representative subsets, of the collection. The genotyping produced 193,988 SNP and 142,522 SilicoDArT markers with an average polymorphic information content of 0.18 and 0.26, respectively. Hierarchical clustering using selected informative markers showed the presence of two and three main clusters using SNP and SilicoDArT markers, respectively, with a cophenetic correction coefficient of 82%. Bayesian population structure analysis also showed the presence of two main subpopulations using both marker types indicating the existence of significant genetic variation in the collection. A representative subset, containing 21 accessions from diverse origins, was developed using the SNP markers. In general, the results revealed substantial genetic diversity in the Rhodes grass collection, and the generated molecular information, together with the developed subset, should help enhance the management, use and improvement of Rhodes grass germplasm in the future.

Rapid metabolic discrimination and prediction of dioscin content from African yam tubers using Fourier transform-infrared spectroscopy combined with multivariate analysis.

To determine whether or not FT-IR spectroscopy could be used for taxonomic and metabolic discrimination of African yam lines, tuber samples from African and Asian yam species were subjected to FT-IR. Most remarkable spectral differences between African and Asian yams were found in the 1750-1700 cm(-1) region, polysaccharide (1200-900 cm(-1)) and protein/amide I and II (1700-1500 cm(-1)) regions of FT-IR spectra. A hierarchical dendrogram based on partial least square-discriminant analysis (PLS-DA) of FT-IR data from 7 African yam species show phylogenetic relationship. In addition, the content of dioscin, a steroidal saponin found in yam tuber, was predicted using a PLS regression model with regression coefficient R(2)=0.7208 indicated that prediction model had average accuracy. Thus, considering these results we suggest that FT-IR combined with multivariate analysis could be applied as a novel tool for metabolic evaluation and high-throughput screening of African yam lines with higher content of dioscin.

Generation and analysis of expressed sequence tags (ESTs) for marker development in yam (Dioscorea alata L.).

BACKGROUND: Anthracnose (Colletotrichum gloeosporioides) is a major limiting factor in the production of yam (Dioscorea spp.) worldwide. Availability of high quality sequence information is necessary for designing molecular markers associated with resistance. However, very limited sequence information pertaining to yam is available at public genome databases. Therefore, this collaborative project was developed for genetic improvement and germplasm characterization of yams using molecular markers. The current investigation is focused on studying gene expression, by large scale generation of ESTs, from one susceptible (TDa 95-0310) and two resistant yam genotypes (TDa 87-01091, TDa 95-0328) challenged with the fungus. Total RNA was isolated from young leaves of resistant and susceptible genotypes and cDNA libraries were sequenced using Roche 454 technology. RESULTS: A total of 44,757 EST sequences were generated from the cDNA libraries of the resistant and susceptible genotypes. Greater than 56% of ESTs were annotated using MapMan Mercator tool and Blast2GO search tools. Gene annotations were used to characterize the transcriptome in yam and also perform a differential gene expression analysis between the resistant and susceptible EST datasets. Mining for SSRs in the ESTs revealed 1702 unique sequences containing SSRs and 1705 SSR markers were designed using those sequences. CONCLUSION: We have developed a comprehensive annotated transcriptome data set in yam to enrich the EST information in public databases. cDNA libraries were constructed from anthracnose fungus challenged leaf tissues for transcriptome characterization, and differential gene expression analysis. Thus, it helped in identifying unique transcripts in each library for disease resistance. These EST resources provide the basis for future microarray development, marker validation, genetic linkage mapping and QTL analysis in Dioscorea species.