Developing a coffee core collection using the principal components score strategy with quantitative data

The genetic organisation of the coffee gene pool was examined at three levels: biogeography, genetic resources and available data. This investigation indicated that a core csllection for coffee should consist of 88 genetic diversity groups of three typs, accordin% to their genetic history and the available genetic knowledge: a Coffcu arabica category, a category contairiing well-studied species such as C. liberica and C. canephoru, and a category with a large number of neglected species. Different strategies were applied to the three categories. After defining ;he diversity groups, tests were conducted, using data obtained for C. liberica, c:! a new method (principal components score strategy) for developing a core collection using quanticative data. The results showed that about half the inertia was obtained when 10% of the 338 genotypes were selected, and 90% of the total inertia was obtained with asample of 50% of these genotypes,