From function to dysfunction: gene ontology and cancer.
In: Student Reports, Summer 2016, Jackson Laboratory
Dr. Mary Dolan and Dr. Judith Blake
The goal of this project was to construct Gene Ontology (GO) ‘slims,’ high level sets of terms to support data aggregation, to provide a new computational pipeline to better understand sets of genes that are dysfunctional in cancer, with a focus on immune system processes and developmental processes. We used functional annotation data in the Mouse Genomic Informatics (MGI) system for mouse, computationally selected the most informative terms, and consulted with biologists to confirm their relevance.
The completed immune system process GO slim of 34 terms covered 1519 out of 1842 total GO immune terms that covered 5257 annotations (73% of annotations to immune system processes overall). Eighty-three excluded terms with direct annotations were grouped as “other immune system process.” The completed developmental system GO slim of 44 terms had the coverage of 3959 out of 5775 total GO immune terms that covered 21219 annotations (84% of annotations to developmental processes overall). Six hundred twenty five excluded terms with direct annotations were grouped as “other developmental processes.” Cancer gene sets of interest were then analyzed using the constructed GO slims to identify which categories of the immune system processes and developmental processes were over or under-represented.
Cha, Yoojin, "From function to dysfunction: gene ontology and cancer." (2016). Summer and Academic Year Student Reports. 2523.