Identificação de sexo e castas em abelhas Apis mellifera utilizando a técnica de Espectroscopia no Infravermelho Próximo

Abstract

The sex and caste differentiation has been understood for some species of bees. In bees of the genus Apis, sex is determined by a cascade of genes including the csd and fem. Already the caste differentiation depends on the amount and quality of nutrition that the larva will receive. The larvae that receive a larger amount and quality of food - royal jelly - throughout its development will lead to the queen, while the larvae that feed fewer and less nutritious food results in workers. Despite these genetic and environmental mechanisms of sex and caste differentiation are well known in Apis mellifera, it is very important the possibility of using advanced techniques that allow the identification and differentiation of individuals at any stage of their development and independent to have information about its cell creates for studies that need on that prior identification. In this sense, the objective of this study was to evaluate the potential of Near-infrared spectroscopy (NIRS) to recognize sexes and castes in different stages of development of Apis mellifera. Were collected males, workers and queens in different stages of development (egg, pre-defecating larvae, defecating larvae and adult newly emerged) and near infrared spectra of each samples were collected for later relationship analysis with gender and caste. In the principal component analysis (PCA) revels that the order of subjects based on two dimensions - the dorsal and lateral, captured 98.29 % of the gross spectral variation for separating the phases. It has been shown by cross-validation, the probability of correctly identifying an individual based on the spectral data was higher than 96 %. The discriminant analyzes indicated a high accuracy of the spectral data to identify sex and caste independently of position (dorsal and lateral). Therefore the Near-infrared spectroscopy (FT- NIRS) showed excellent performance, standing out as a promising tool to distinguish gender and caste in Apis mellifera bees.