Association Mapping for Sugarcane Quality Traits at Three Harvest Times

• Published in: Sugar tech : an international journal of sugar crops & related industries 2022-04, Vol.24 (2), p.448-462
• Main Authors: Coutinho, Alisson Esdras, da Silva, Marcel Fernando, Perecin, Dilermando, Carvalheiro, Roberto, Xavier, Mauro Alexandre, de Andrade Landell, Marcos Guimarães, Pinto, Luciana Rossini
• Format: Article
• Language: English
• Subjects: Agriculture; Amplified fragment length polymorphism; Analysis; Association analysis; Associations, institutions, etc; Bayesian analysis; Biomass; Biomedical and Life Sciences; Cluster analysis; Clustering; Gene mapping; Gene polymorphism; Genotypes; Life Sciences; Mapping; Markers; Phenotypic variations; Polymorphism; Quality control; Research Article; Societies; Spring; Spring (season); Sugar; Sugarcane; Vector quantization
• ISSN: 0972-1525; 0974-0740; 0972-1525
• DOI: 10.1007/s12355-021-01056-5

Sugarcane is a major crop cultivated globally for sugar and bioenergy production, with increasing relevance as a biomass source. Association mapping studies, in turn, identify markers associated with target traits that may assist genotype selection. Here, we used a diversity panel comprising 100 sugarcane genotypes to investigate the clustering patterns using phenotypic data, i.e., sugar content (Pol%Cane) and fiber, and Amplified Fragment Length Polymorphism (AFLP) markers, and perform association mapping to identify marker–trait associations (MTAs) that are consistent across harvesting times, i.e., autumn, winter and spring harvest, and across 2 years for the spring harvest. The K-means clustering of phenotypic and genotypic data revealed discontinuities among genotypes, indicating a distribution into two groups with high intra- and inter-group diversity. A subset of 640 AFLP markers across 93 genotypes was selected after quality control and subjected to association analysis via the Bayes C method. Different sets of MTAs were detected across harvest times for each trait, with each set collectively explaining 68.44–99.75% of the phenotypic variation (R 2 ). The detection in more than one harvest time was observed for 32 and 12 MTAs for Pol%Cane and fiber, respectively, while none was detected in all the harvest times. The most important markers selected by the stepwise process individually explained 1.61–67.55% of the phenotypic variation, with highlights for the MTAs with high individual R 2 values and/or detection in two or more harvest times with consistent positive effects, which may correspond to new genomic regions associated with Pol%Cane and fiber and should be further investigated.