A review on Gene Ontology evaluations
Borja Pitarch, Mónica Chagoyen, Juan A. G. Ranea, Florencio Pazos
Abstract
The analysis of the large and heterogeneous datasets that characterize modern biology demands systems capable of representing biological knowledge in a formal, standardized manner. The most widely used structured vocabulary in molecular biology and biomedicine is the Gene Ontology (GO). Although employed in many different data analysis tasks, its main application is in the functional annotation of gene products and the subsequent functional enrichment analysis of large gene/protein datasets derived from omics experiments. This key position in modern biology means the GO is subject to continuous scrutiny and improvement, not only by its developers but by the community in general. GO's vocabulary, structure, and gene annotations have been evaluated from multiple points of view, revealing important insights into how we organize our biological knowledge, the associated problems, and strategies for overcoming them. These evaluations range from technical analyses of its ontological architecture to assessments of its effectiveness in supporting biological research. This process ensures that this essential resource is updated and adapted to the new challenges posed by modern biology. In this review, we summarize some key evaluations that the GO has undergone over time.