Plant breeding can be broadly defined as alterations caused in plants as a result of their use by humans, ranging from unintentional changes resulting from the advent of agriculture to the application of molecular tools for precision breeding. The vast diversity of breeding methods can be simplified into three categories: (i) plant breeding based on observed variation by selection of plants based on natural variants appearing in nature or within traditional varieties; (ii) plant breeding based on controlled mating by selection of plants presenting recombination of desirable genes from different parents; and (iii) plant breeding based on monitored recombination by selection of specific genes or marker profiles, using molecular tools for tracking within-genome variation. The continuous application of traditional breeding methods in a given species could lead to the narrowing of the gene pool from which cultivars are drawn, rendering crops vulnerable to biotic and abiotic stresses and hampering future progress. Several methods have been devised for introducing exotic variation into elite germplasm without undesirable effects. Cases in rice are given to illustrate the potential and limitations of different breeding approaches.

Some species of Trichoderma have successfully been used in the commercial biological control of fungal pathogens, e.g., Sclerotinia sclerotiorum, an economically important pathogen of common beans (Phaseolus vulgaris L.). The objectives of the present study were (1) to provide molecular characterization of Trichoderma strains isolated from the Brazilian Cerrado; (2) to assess the metabolic profile of each strain by means of Biolog FF Microplates; and (3) to evaluate the ability of each strain to antagonize S. sclerotiorum via the production of cell wall-degrading enzymes (CWDEs), volatile antibiotics, and dual-culture tests. Among 21 isolates, we identified 42.86% as Trichoderma asperellum, 33.33% as Trichoderma harzianum, 14.29% as Trichoderma tomentosum, 4.76% as Trichoderma koningiopsis, and 4.76% as Trichoderma erinaceum. Trichoderma asperellum showed the highest CWDE activity. However, no species secreted a specific group of CWDEs. Trichoderma asperellum 364/01, T. asperellum 483/02, and T. asperellum 356/02 exhibited high and medium specific activities for key enzymes in the mycoparasitic process, but a low capacity for antagonism. We observed no significant correlation between CWDE and antagonism, or between metabolic profile and antagonism. The diversity of Trichoderma species, and in particular of T. harzianum, was clearly reflected in their metabolic profiles. Our findings indicate that the selection of Trichoderma candidates for biological control should be based primarily on the environmental fitness of competitive isolates and the target pathogen.

Common bean is a legume of social and nutritional importance as a food crop, cultivated worldwide especially in developing countries, accounting for an important source of income for small farmers. The availability of the complete sequences of the two common bean genomes has dramatically accelerated and has enabled new experimental strategies to be applied for genetic research. DArTseq has been widely used as a method of SNP genotyping allowing comprehensive genome coverage with genetic applications in common bean breeding programs. Using this technology, 6286 SNPs (1 SNP/86.5 Kbp) were genotyped in genic (43.3%) and non-genic regions (56.7%). Genetic subdivision associated to the common bean gene pools (K = 2) and related to grain types (K = 3 and K = 5) were reported. A total of 83% and 91% of all SNPs were polymorphic within the Andean and Mesoamerican gene pools, respectively, and 26% were able to differentiate the gene pools. Genetic diversity analysis revealed an average H E of 0.442 for the whole collection, 0.102 for Andean and 0.168 for Mesoamerican gene pools (F ST = 0.747 between gene pools), 0.440 for the group of cultivars and lines, and 0.448 for the group of landrace accessions (F ST = 0.002 between cultivar/line and landrace groups). The SNP effects were predicted with predominance of impact on non-coding regions (77.8%). SNPs under selection were identified within gene pools comparing landrace and cultivar/line germplasm groups (Andean: 18; Mesoamerican: 69) and between the gene pools (59 SNPs), predominantly on chromosomes 1 and 9. The LD extension estimate corrected for population structure and relatedness (r2 SV) was ~ 88 kbp, while for the Andean gene pool was ~ 395 kbp, and for the Mesoamerican was ~ 130 kbp. For common bean, DArTseq provides an efficient and cost-effective strategy of generating SNPs for large-scale genome-wide studies. The DArTseq resulted in an operational panel of 560 polymorphic SNPs in linkage equilibrium, providing high genome coverage. This SNP set could be used in genotyping platforms with many applications, such as population genetics, phylogeny relation between common bean varieties and support to molecular breeding approaches.

The species of T. harzianum are well known for their biocontrol activity against plant pathogens. However, few studies have been conducted to further our understanding of its role as a biological control agent against S. sclerotiorum, a pathogen involved in several crop diseases around the world. In this study, we have used RNA-seq and quantitative real-time PCR (RT-qPCR) techniques in order to explore changes in T. harzianum gene expression during growth on cell wall of S. sclerotiorum (SSCW) or glucose. RT-qPCR was also used to examine genes potentially involved in biocontrol, during confrontation between T. harzianum and S. sclerotiorum.

Estimation of allele dosage, using genomic data, in autopolyploids is challenging and current methods often result in the misclassification of genotypes. Some progress has been made when using SNP arrays, but the major challenge is when using next generation sequencing data. Here we compare the use of read depth as continuous parameterization with ploidy parameterizations in the context of genomic selection (GS). Additionally, different sources of information to build relationship matrices were compared. A real breeding population of the autotetraploid species blueberry (Vaccinium corybosum), composed of 1,847 individuals was phenotyped for eight yield and fruit quality traits over two years. Continuous genotypic based models performed as well as the best models. This approach also reduces the computational time and avoids problems associated with misclassification of genotypic classes when assigning dosage in polyploid species. This approach could be very valuable for species with higher ploidy levels or for emerging crops where ploidy is not well understood. To our knowledge, this work constitutes the first study of genomic selection in blueberry. Accuracies are encouraging for application of GS for blueberry breeding. GS could reduce the time for cultivar release by three years, increasing the genetic gain per cycle by 86% on average when compared to phenotypic selection, and 32% when compared with pedigree-based selection. Finally, the genotypic and phenotypic data used in this study are made available for comparative analysis of dosage calling and genomic selection prediction models in the context of autopolyploids.

Association analysis was applied to a panel of accessions of Embrapa Rice Core Collection (ERiCC) with 86 SSR and field data from two experiments. A clear subdivision between lowland and upland accessions was apparent, thereby indicating the presence of population structure. Thirty-two accessions with admixed ancestry were identified through structure analysis, these being discarded from association analysis, thus leaving 210 accessions subdivided into two panels. The association of yield and grain-quality traits with SSR was undertaken with a mixed linear model, with markers and subpopulation as fixed factors, and kinship matrix as a random factor. Eight markers from the two appraised panels showed significant association with four different traits, although only one (RM190) maintained the marker-trait association across years and cultivation. The significant association detected between amylose content and RM190 was in agreement with previous QTL analyses in the literature. Herein, the feasibility of undertaking association analysis in conjunction with germplasm characterization was demonstrated, even when considering low marker density. The high linkage disequilibrium expected in rice lines and cultivars facilitates the detection of marker-trait associations for implementing marker assisted selection, and the mining of alleles related to important traits in germplasm.