QTL mapping in population originated from a bi-parental commercial cross of sugarcane

Author(s)

Thiago Willian Almeida Balsalobre

• Document type: Doctoral Thesis
• Press: Campinas, SP.
• Institution: Universidade Estadual de Campinas (UNICAMP). Instituto de Biologia
• Defense date: 2016-11-29

Examining board members

Anete Pereira de Souza; Renato Vicentini; Marcelo Menossi; Roberto Giacomini Chapola; João Ricardo Bachega Feijó Rosa

• Advisor: Monalisa Sampaio Carneiro; Anete Pereira de Souza

Abstract

Sugarcane is a renewable source of energy and with potential for expansion. The genetic complexity of sugarcane due to its high level of ploidy and aneuploidy, together with the quantitative nature of most agronomic traits, has hindered currently the achievement of high rates of genetic gain through conventional breeding of this crop. The development of molecular markers and construction of genetic maps can be helpful to establish strategies in breeding programs and increase the efficiency of the selection process and accelerate the development of new cultivars. Therefore, the purpose of this thesis was to construct an integrated genetic map in sugarcane and identify genomic regions that control traits of interest by QTL mapping (Quantitative Trait Loci). To achieve these goals we used an F1 segregating population of sugarcane with 153 individuals from bi-parental commercial cross between cultivars SP80-3280 and RB835486. The field trial was carried out in two locations, Araras-SP and Ipaussu-SP, and the experimental design consisted of an augmented randomized incomplete block, which was fully replicated three times. The phenotypic evaluations were performed over three years (2011, 2012 and 2013). We applied a mixed model approach for the analysis of phenotypic traits related to yield components and resistance to brown rust. The severity data of brown rust was analyzed by generalized linear mixed model. Heritability estimates were high, ranging from 0.78 (stalk height) to 0.92 (stalk diameter), and the brown rust severity analysis showed that 66% of the clones have at least 90% probability of being resistant to disease. To construct an integrated genetic map were used molecular markers microsatellites (Simple Sequence Repeats, SSR), TRAP (Target Target Region Amplification Polymorphism), as well as SNPs (Single Nucleotide Polymorphims) and indels derived from the GBS protocol (Genotyping-by-Sequencing). To GBS-based markers discovery were used four pseudo-references: sorghum genome (Sorghum bicolor), methyl-filtered genome of sugarcane, transcriptome of sugarcane (RNAseq) and sequences of SUCEST project. The ploidy and allelic dosage of each bi-allelic locus was estimated by SUPERMASSA software. Using Onemap software (v. 2.0-4) and employing LOD> 9.0 and recombination fraction <0.10, a total of 993 markers in single dose were mapped. These markers were distributed throughout 223 linkage groups, which were clustered in 18 homo(eo)logous groups. The total length of the map was 3,682.04 cM with an average marker density of 3.70 cM. Using composite interval mapping (Composite Interval Mapping, CIM) were mapped seven QTLs considering four of the 11 phenotypic traits evaluated. The results suggest the presence of a stable QTL across locations to soluble solid content (BRIX) and sucrose content of the cane (POL%C). Furthermore, QTLs to BRIX and fiber content (FIB) traits had associated markers with candidate genes, which had great potential for validation and future use for molecular breeding in sugarcane. This study is the first to report the use of GBS for large-scale variant discovery and genotyping of a population in sugarcane with posterior analysis to composite interval mapping (AU)