04226naa a2200229 a 450000100080000000500110000800800410001910000210006024501750008126000090025652034590026565300190372465300110374365300160375470000190377070000210378970000190381070000170382970000170384670000200386377301130388311281612018-12-17 2018 bl uuuu u00u1 u #d1 aSGANZERLA, W. G. aAcca Sellowiana accesses discrimination by nutritional composition, phenolic compounds and antioxidant activity using Principal Component Analysis.h[electronic resource] c2018 aFeijoa [Acca Sellowiana (O. Berg) Burret] is a native fruit from Southern region of Brazil, but there are few feijoa planting on a commercial scale. Researches aim to advance the knowledge about the species, with regard to genetic improvement. Thus, genetic improvement has been performed since 1986 for Company of Agricultural Research and Rural Extension of Santa Catarina (EPAGRI), with this plant to improve the pulp yield, nutritional quality and production regularity (Santos, Ciotta and Nodari, 2017). Nowadays, four cultivars from feijoa developed in Brazil were launched in the national and international market. Objective: The aim of this study was to discriminate seven accesses of feijoa by nutritional composition, phenolic compounds and antioxidant activity using Principal Component Analysis (PCA). Methods: The samples are originating from the genotype evaluation experiment (genetic enhancement) carried out by EPAGRI (Lages, SC) since 1998 with native progenies from Brazil (Santa Catarina and Rio Grande do Sul State), Uruguay and New Zealand. The material collected in this study comes from accesses selected and planted in 2008 (Santos et al., 2011). Seven accesses were harvest in the physiological maturation point. Feijoa pulp and husk were manually separated and analyzed. On feijoa pulp, it was quantified the content of moisture, ashes, lipids, proteins, carbohydrates, pH, Total Acidity (TA) and Total Soluble Solids (TSS). The same analyses were carried out on feijoa husk, except pH, TA and TSS. Total Phenolic Compounds (TPC), and antioxidant activity (DPPH, ABTS and FRAP assay) were quantified on feijoa pulp and husk extracts. Extracts were performed according to Sganzerla et al. (2018) using hydroethanolic solution (70 ??GL, 1:10 m/v). Multivariate analysis (pulp), analysis of variance and Tukey?s test (p<0.05) (pulp and husk) were used to data interpretation, using Statistica?? 7.0 software. Results and Discussion: Accesses of feijoa (pulp and husk) presented statistical difference on moisture, ashes and carbohydrates. Moreover, pH, TA and TSS were statistical difference on feijoa pulp. TPC and antioxidant activity (DPPH, ABTS and FRAP assay) were higher on husk compared to pulp, and were statistical difference between accesses (p<0.05). Evaluating the Pearson?s correlation coefficients, on feijoa pulp, TPC and antioxidant activity presented a higher correlation coefficients (p<0.05), but lipids did not present significative correlation coefficient with any variable. Variables with significative correlations were selected to multivariate analysis (PCA). The variables projections on the factor-plane were explained in 78% (PC1 + PC2), and the accesses of feijoa pulp were discriminated in four groups (1st group: access 5; 2nd group: accesses 6; 3rd group: access 1 and 7; and 4th group: accesses 2, 3 and 4). The discrimination of the access 5 can be explained (p<0.05) by higher content of ashes, carbohydrates, TSS, TPC and antioxidant activity (DPPH, ABTS and FRAP assay). These variables were projected in the same region of the variable graphic, which resulted in an isolated access (access 5). Conclusion: It was possible to discriminate accesses of feijoa using parameters of nutritional composition, phenolic compounds and antioxidant activity using PCA. The access 5 of the feijoa genetic improvement presented higher nutritional value, phenolic compounds and antioxidant activity. aclassification aFeijoa aFood design1 aAZEVEDO, M. S.1 aFERRAREZE, J. P.1 aKOMATSU, R. A.1 aNUNES, M. R.1 aCOSTA, M. D.1 aVEECK, A. P. L. tIn: INTERNATIONAL CONGRESS ON BIOACTIVE COMPOUNDS, 1., 2018, Campinas. Abstracts... Campinas: Unicamp, 2018.