Automated Predictions by a Factor Graph GO Annotation in Pacu

MARIANO TORRES MANNO; SPETALE FLAVIO EZEQUIEL; BULACIO PILAR; JOAQUÍN EZPELETA; TAPIA ELIZABETH; SILVIA ARRANZ; VANINA VILLANOVA; KRSTICEVIC FLAVIA

Buenos Aires

Simposio; 2nd Latin America Student Council Symposium; 2016

Gene Ontology is a hierarchical controlled vocabulary used to describe gene func on. Protein-coding genes are generally annotated through standard methods that commonly rely on sequence similarity or protein signature searches. Most of the gene func onal characteriza on in model organisms remains without a complete GO annota on. The worst-case scenario is for non-model organisms, where genome reference is not available and gene novel es with unknown func on are expected. As a consequence, alterna ve computa onal method for automated gene annota on can be useful. In this work a classifica on method based on factor graph GO annota on (FGGA) is considered to enrich the coding gene annota on in non-model organisms. Concerning characteriza on methods of individual protein sequences in terms of 453 input features of the physicochemical and the secondary structure type. Training was performed with 3573 protein sequences from the zebrafish model organism with similarity and experimental GO code evidences. The aim of this work was to evaluate the ability of FGGA predic ons to obtain more specific GO terms in a set of 441 pair paralogous genes in the non-model organism Pacú (Piaractus mesopotamicus). To recover most specific and confident FGGA predic ons a cut threshold of 0.95 for leaf nodes was set for the analysis of predicted graphs. Addi onally, to extend biological knowledge in the Pacú, 5 genes with unknown func on were considered. Finally, par cular a en on was given to specific genes involved in growth and developmental process in Pacú. These target genes could be useful for markers iden fica on for future breeding program.