The sensory properties of cacao beans are linked to the chemical composition of the seeds, and both characteristics are the partial results of its allelic composition. Therefore, it is useful to search for molecular markers associated with these traits. We perform multiple regression analysis to associate previously generated data of alleles generated with 12 SSR (of cultivated cacao trees) with data obtained from chemical and sensory characterization (of beans) of plants grown in the southern region from Mexico. When the association was significant, the mathematical models for predictive purposes were proposed. All phenotypic traits evaluated showed equations with setting values R2 > 0.5. All chemical characters tested have a significant association with at least two alleles (P < 0.05). In addition, the fat content was associated with six molecular markers (mTcCIR03209, mTcCIR12188, mTcCIR19286, mTcCIR07150, mTcCIR19310). The most common allele was mTcCIR12188, which was associated with the contents of eicosanoic acid, moisture, fat and total polyphenols content. The mTcCIR28362 allele is associated with sensory characters bitterness, musty odor, and roasted odor. These alleles could be useful as molecular markers of chemical and sensory characteristics of cacao samples.
Keywords: Eicosanoic acid; Fat content; Flavor; Odor; Polyphenols.