Iron defi ciency chlorosis (IDC) is a signifi cant yield-limiting problem in several major soybean [Glycine max (L.) Merr.] production regions in the United States. Soybean plants display a variety of symptoms that range from a slight yellowing of the leaf to interveinal chlorosis, to stunted growth that reduces yield. The objective of this analysis was to employ single nucleotide polymorphism (SNP)-based genome-wide association mapping to uncover genomic regions associated with IDC tolerance. A number of the loci discovered here mapped at or near previously discovered IDC quantitative trait loci (QTL). A total of 15 genes known to be involved in iron metabolism mapped in the vicinity (<500 kb) of signifi cant markers in one or both populations.

Iron deficiency chlorosis (IDC) is a yield limiting problem in soybean (Glycine max (L.) Merr) production regions with calcareous soils. Genome-wide association study (GWAS) was performed using a high density SNP map to discover significant markers, QTL and candidate genes associated with IDC trait variation. A stepwise regression model included eight markers after considering LD between markers, and identified seven major effect QTL on seven chromosomes. Twelve candidate genes known to be associated with iron metabolism mapped near these QTL supporting the polygenic nature of IDC. A non-synonymous substitution with the highest significance in a major QTL region suggests soybean orthologs of FRE1 on Gm03 is a major gene responsible for trait variation. NAS3, a gene that encodes the enzyme nicotianamine synthase which synthesizes the iron chelator nicotianamine also maps to the same QTL region. Disease resistant genes also map to the major QTL, supporting the hypothesis that pathogens compete with the plant for Fe and increase iron deficiency. The markers and the allelic combinations identified here can be further used for marker assisted selection.

Variation in common bean (Phaseolus vulgaris L.) was investigated by sequencing intron 1 of the dihydroflavonol 4-reductase (DFR) gene for 92 genotypes that represent both landraces and cultivars. We were also interested in determining if introns provide sufficient variation for genetic diversity studies and if the sequence data could be used to develop allele-specific primers that could differentiate genotypes using a standard PCR assay. Sixty-nine polymorphic sites were observed. Nucleotide variation (pi/bp) was 0.0481, a value higher than that reported for introns from other plant species. Tests for significant deviation from the mutation drift model were positive for the population as a whole, the cultivar and landrace subsets, and the Middle American landrace set. Significant linkage disequilibrium extended about 300 nucleotides. Twenty haplotypes were detected among the cultivated genotypes. Seven recombination events were detected for the whole population, and six events for the landraces. Recombination was not observed among the landraces within either the Middle American or Andean gene pools. Evidence for hybridization between the two gene pools was discovered. Five allele-specific primers were developed that could distinguish 56 additional genotypes. The allele-specific primers were used to map duplicate DFR genes on linkage group B8.

Crop species have been deeply affected by the domestication process, and there have been many efforts to identify selection signatures at the genome level. This knowledge will help geneticists to better understand the evolution of organisms, and at the same time, help breeders to implement successful breeding strategies. Here, we focused on domestication in the Mesoamerican gene pool of Phaseolus vulgaris by sequencing 49 gene fragments from a sample of 45 P. vulgaris wild and domesticated accessions, and as controls, two accessions each of the closely related species Phaseolus coccineus and Phaseolus dumosus. An excess of nonsynonymous mutations within the domesticated germplasm was found. Our data suggest that the cost of domestication alone cannot explain fully this finding. Indeed, the significantly higher frequency of polymorphisms in the coding regions observed only in the domesticated plants (compared to noncoding regions), the fact that these mutations were mostly nonsynonymous and appear to be recently derived mutations, and the investigations into the functions of their relative genes (responses to biotic and abiotic stresses), support a scenario that involves new functional mutations selected for adaptation during domestication. Moreover, consistent with this hypothesis, selection analysis and the possibility to compare data obtained for the same genes in different studies of varying sizes, data types, and methodologies allowed us to identify four genes that were strongly selected during domestication. Each selection candidate is involved in plant resistance/tolerance to abiotic stresses, such as heat, drought, and salinity. Overall, our study suggests that domestication acted to increase functional diversity at target loci, which probably controlled traits related to expansion and adaptation to new agro-ecological growing conditions.

Since common bean is the most important legume crop for human consumption around the world, bean breeders are challenged to increase production of beans while facing new problems like climate change. Guatemalan climbing beans have been suggested to represent race Guatemala, a newly identified race in the Middle American gene pool that may represent an untapped source of alleles for bean improvement. This study confirmed the existence of race Guatemala in the Middle American gene pool and its differentiation from other races. The low population structure found within these Guatemalan beans also makes this population ideal for discovery of candidate genes for important traits. We demonstrate that the Guatemalan population was useful to provide candidate genes for previously reported genetic factors like the V gene for flower color, and the Asp gene for seed coat luster. The important relationship between flowering time and altitudinal adaptation of beans was also emphasized. I would also like to acknowledge Dr. Samira Mafi Moghaddam, Rian Lee and Dr. Atena Oladzad from the Dry Bean Genomics Lab at NDSU for their instruction and help on each step of this project. Without their guidance, this project would have never been possible to conclude.