This study aimed at the monitoring of the apricot (Prunus armeniaca L.) ripening progression through the expression analysis of 25 genes related to fruit quality traits in nine cultivars with great differences in fruit color and ripening date. The level of pigment compounds, such as anthocyanins and carotenoids, is a key factor in food taste, and is responsible for the reddish blush color or orange skin and flesh color in apricot fruit, which are desirable quality traits in apricot breeding programs. The construction of multiple linear regression models to predict anthocyanins and carotenoids content from gene expression allows us to evaluate which genes have the strongest influence over fruit color, as these candidate genes are key during biosynthetic pathways or gene expression regulation, and are responsible for the final fruit phenotype. We propose the gene CHS as the main predictor for anthocyanins content, CCD4 and ZDS for carotenoids content, and LOX2 and MADS-box for the beginning and end of the ripening process in apricot fruit. All these genes could be applied as RNA markers to monitoring the ripening stage and estimate the anthocyanins and carotenoids content in apricot fruit during the ripening process.
Keywords: Prunus armeniaca L.; breeding; fruit quality; metabolomics; multiple linear regression; predictive models; ripening; transcriptomics.