Genstat and QTL analysis for Sugarcane

Quantitative trait loci (QTL) mapping in sugarcane is more challenging than other species because the combination of the ploidy level and outbred parents makes the genetic pattern in the segregating progenies more complex. In addition to its genetic complexity, sugarcane is a perennial crop, in which individuals are usually harvested in multiple years. Repeated measures are obtained for various traits across different locations along harvest years. Consistent QTL across different environments (any combination of location and year) is very useful for marker-assisted selection breeding programs.

To identify stable QTL, the effect of genotype-by-environment (G x E) interaction should be considered and deployed in the mixed models as well as QTL-by-environment (QTL x E) interaction. Pastina et al. (2012) described the model for QTL mapping in sugarcane based on mixed models for single-trait multi-environment data (emphasizing correlations between location and year). The results showed stable QTLs that can be distinguished from environment-sensitive QTLs. Margarido et al. (2015) also reported the analysis of QTL and QTL x E by fitting mixed models with multiple QTLs, with appropriate modeling of heterogenous genetic (co)variance between traits, locations and years (multi-trait multi-environment). The results of these studies provide useful information on the genetic basis of quantitative variation in sugarcane and the genetic relation between traits.

Genstat is used in the studies to detect QTLs for multi-trait multi-environment trials based on mixed models. The steps include;

  • Identifying the best variance-covariance model for the phenotypic data
    • A number of models can be fitted and compared based on their AIC (Akaike Information Criterion) and BIC (Bayesian Information Criterion) values.
    • Using the best model provides reliable predictions of genotype performance across environments.
  • Searching for QTLs using single marker analysis and interval mapping
    • Genetic predictors can be calculated from molecular marker information and used as explanatory variables in the QTL models.
  • Fitting a final multi-QTL model to estimate QTL effects
    • A joint test for QTL and/or QTL x E interaction effects can be performed and a final model will be used to estimate QTL effects for each trait, in each location and harvest year.



Margarido GRA, Pastina MM, Souza AP and Garcia AAF. 2015. Multi-trait multi-environment quantitative trait loci mapping for a sugarcane commercial cross provides insights on the inheritance of important traits. Mol Breeding. 35:175.

Pastina MM, Malosetti M, Gazaffi R, Mollinari M, Margarido GRA, Oliveira, Pinto LR, Souza AP, Eeuwijk FA and Garcia AAF. 2012. A mixed model QTL analysis for sugarcane multiple-harvest-location trial data. Theor Appl Genet. 124:835-849.