DATA MINING AND SYSTEMS BIOLOGY FOR IDENTIFYING KEY GENES INVOLVED IN CITRUS QUALITY

Abstract

Quality in citrus is mainly characterized by fruit and juice colour, fruit and skin size, juice percent, total soluble solids, titrable acidity, and carotenoid/flavonoid contents. Moreover, studies of biosynthetic pathway of the metabolites/proteins involved in quality at transcriptional and translational levels may give relevant information for subsequent functional studies and quality improvement. Data mining of ESTs from HarvEST database allowed the selection of 17 cDNA libraries from albedo, flavedo, peel, pulp and juice sac of different orange, mandarin, clementine and grapefruit varieties. In order to select key genes involved in quality we used systems biology that offers mathematical tools that include the analysis of the structure, clustering and centralities of the network. In order to have information regarding physical protein-protein interactions (PPPI) from citrus sequences, orthologous sequences of Arabidopsis thaliana were used (BLASTX; reciprocal BLASTP). Literature data mining was performed, and PPPI network design was obtained using the Cytoscape software. The interactome networks thus obtained were analyzed with MCODE. Gene ontology clustering analysis was performed using BiNGO. Specific algorithms were applied to identify modules and central nodes within the citrus libraries associated network. The obtained results will be used as a guideline to select specific genes/proteins from citrus for further functional studies as gene expression or plant transformation. (Resume)