Discrimination of Genealogical Groups of Arabica Coffee by the Chemical Composition of the Beans

Larissa O Fassio,Rosemary G F A Pereira,Marcelo R Malta,Gladyston R Carvalho,Ackson D Silva,Gilberto R Liska,Vany P Ferraz,Adriene W Pedrosa,Antônio A Pereira

doi:10.5539/jas.v11n16p141

Abstract

This work aimed to characterize and discriminate genealogical groups of coffee as to the chemical composition of the grains through the model created by PLS-DA method. 22 accessions of Coffea arabica, from the Active Germplasm Bank of Minas Gerais, were divided into groups according to the genealogical origin. Samples of ripe fruits were harvested selectively and processed by the wet method, to obtain pulped coffee beans, with 11% (b.u.) of water content. The raw beans were assessed as to the content of polyphenols, total sugars, total lipids, protein, caffeine, sucrose, and fatty acids. The data were submitted the chemometric analysis, PCA and PLS-DA. The results of PLS-DA identified the variables which most influence the classification of genealogical groups and possible chemical markers to accessions processed by the pulped method. The sucrose content was an important marker for the Exotic accession group. However, the content of polyphenols has been identified as a marker for the group Tymor Hybrid, and the caffeine for the bourbon group. The different fatty acids have been identified as markers for all genealogical groups, at different levels. The model PLS-DA is effective in discriminating genealogical groups from the chemical composition of the beans.

Highlights

In recent years, due to the growing market for quality coffee, producers have sought for cultivars with features that meet the standard of quality expected
The experiment was developed in the agricultural year of 2016/2017 with samples of raw beans of 22 accessions of Coffea arabica divided into groups according to the genealogical origin (Table 1)
The PC1 shows that there is a trend in the accessions MG0223, MG0303, and MG0228 to present a higher content of linolenic fatty acid and the accession MG0235 for linoleic fatty acid

Summary

Introduction

Due to the growing market for quality coffee, producers have sought for cultivars with features that meet the standard of quality expected. The challenge of breeders and researchers is to develop cultivars with desired agronomic characteristics and with a high-quality beverage (Bertrand et al, 2008). The quality of the drink can be evaluated by the attributes of flavor and aroma produced during the roasting process from chemical compounds in the raw bean. The chemical composition of the beans can vary based on certain factors such as species, environment, processing, and drying, and storage (Borém et al, 2016; Cheng et al, 2016; Clemente et al, 2015; Fassio et al, 2016). It is known that the genetic factor is one of the most important and that several compounds in the coffee bean are genetically controlled, such as caffeine, lipids, fatty acids, trigonelline, chlorogenic acid, sugar and proteins (Scholz et al, 2013; Taveira et al, 2014). The use of multivariate statistical methods, such as principal component analysis (PCA) and partial least-squares jas.ccsenet.org

Objectives

Methods

Results