SELECTION OF COVARIATES IN A LOGISTIC REGRESSION MODEL FOR THE PREDICTION OF RESISTANCE TO RICE BLAST
Main Article Content
Abstract
Rice (Oryza sativa L.) has been one of the most consumed foods on the planet, with economic and social importance. Diseases, mainly blast, caused by the fungus Pyricularia oryzae, are limiting factors for the production of rice. The present work aimed to select covariables that can influence resistance to rice blast, using the selection strategy proposed by Collett. Logistic regression models were adjusted to predict disease resistance, using the ROC curve to assess the predictive capacity. The data used were obtained from a population of 413 plants, with phenotypic information collected in 82 countries and classified into five subpopulations. The research found that, out of over fifteen variables embedded to assess the disease, only three revealed to be relevant for the final adjusted model, namely: width of flag leaf (V4), the mean number of primary panicle branches (V8) and the amount of amylose from ground grains (V15). The variable V4 presented the most significant influence on disease resistance. Additionally, for each unit increase in V4, V8 and V15, it is expected to obtain 279.3, 31.9 and 9.4% increases, respectively, in the probability of resistance to rice blast.
Article Details
This work is licensed under a Creative Commons Attribution 4.0 International License.
Authors who publish with this journal agree to the following terms:
- Authors retain copyright and grant the journal right of first publication with the work simultaneously licensed under a Creative Commons Attribution License that allows others to share the work with an acknowledgement of the work's authorship and initial publication in this journal.
- Authors are able to enter into separate, additional contractual arrangements for the non-exclusive distribution of the journal's published version of the work (e.g., post it to an institutional repository or publish it in a book), with an acknowledgement of its initial publication in this journal.
- Authors are permitted and encouraged to post their work online (e.g., in institutional repositories or on their website) prior to and during the submission process, as it can lead to productive exchanges, as well as earlier and greater citation of published work (See The Effect of Open Access).