Summary
Coffee species such as
Coffea canephora
P. (Robusta) and
C. arabica
L. (Arabica) are important cash crops in tropical regions around the world.
C. arabica
is an allotetraploid (2
n
= 4
x
= 44) originating from a hybridization event of the two diploid species
C. canephora
and
C. eugenioides
(2
n
= 2
x
= 22). Interestingly, these progenitor species harbour a greater level of genetic variability and are an important source of genes to broaden the narrow Arabica genetic base. Here, we describe the development, evaluation and use of a single‐nucleotide polymorphism (
SNP
) array for coffee trees. A total of 8580 unique and informative
SNP
s were selected from
C. canephora
and
C. arabica
sequencing data, with 40% of the
SNP
located in annotated genes. In particular, this array contains 227 markers associated to 149 genes and traits of agronomic importance. Among these, 7065
SNP
s (~82.3%) were scorable and evenly distributed over the genome with a mean distance of 54.4 Kb between markers. With this array, we improved the Robusta high‐density genetic map by adding 1307
SNP
markers, whereas 945
SNP
s were found segregating in the Arabica mapping progeny. A panel of
C. canephora
accessions was successfully discriminated and over 70% of the
SNP
markers were transferable across the three species. Furthermore, the canephora‐derived subgenome of
C. arabica
was shown to be more closely related to
C. canephora
accessions from northern Uganda than to other current populations. These validated
SNP
markers and high‐density genetic maps will be useful to molecular genetics and for innovative approaches in coffee breeding.