Ambiguous genetic relationships among coconut (Cocos nucifera L.) cultivars: the effects of outcrossing, sample source and size, and method of analysis

Abstract

A prior analysis of eight coconut cultivars with 15 microsatellite (SSR) markers drew unexpected relationships between two of the out-crossing tall cultivars evaluated: ‘Atlantic Tall’ and ‘Panama Tall’. We further investigated the relationships between these eight cultivars by increasing the number of individuals studied (particularly for ‘Atlantic Tall’ and ‘Panama Tall’), by including 28 more molecular markers, and by adding two other cultivars to our analysis. Our results show that five to ten coconut individuals do not represent a dependable sample to withdraw conclusions regarding cultivar/variety relationships, particularly when studying out-crossing genotypes. As suggested in the prior study, a high level of hybridization was observed between the ‘Atlantic Tall’ and ‘Panama Tall’ cultivars. However, at this time we were able to identify distinct groups for each one of these two cultivars. The two clustering methods used (Neighbor Joining, NJ and Unweighted Pair Group Method with Arithmetic mean, UPGMA) produced dendrograms that resolved contrasting cultivar relationships, especially for the ‘Atlantic Tall’ and ‘Panama Tall’ cultivars. We discuss the implications of our results in regard to current scenarios of coconut domestication and future considerations when assessing genetic relationships among different varieties.