Overview of Aims and Deliverables

From GO Wiki
Jump to navigation Jump to search

The Gene Ontology (GO) is a ubiquitous tool in biomedical research; widely incorporated, as an integral component, into numerous bioinformatics tools and workflows. With new technologies generating ever more insights into fundamental biology, the GO project will continue to build on its existing strengths: bringing innovative approaches to capturing computable representations of experimental results; inferring additional functional roles based upon the principled application of evolutionary relationships; and assuring the GO resource provides the highest accuracy and most up-to-date functional information available. Our new tooling will not only increase the productivity of our GO biocurators, it will also offer biocurators more flexibility for expressing the fine details and nuances of the biology. GO biocurators will simultaneously be able to update the ontology in real time with the new annotation tool, reflecting the biology manifested in the ontology in a much more natural, intuitive process. Perhaps most strikingly, our Common Annotation Framework and increasingly intuitive tools and processes enable a direct migration path to community annotation, in anticipation of which GO biocurators will be putting considerable effort into generating user documentation and training materials. And, of course, the GO Consortium prioritizes the quality of this resource, and the bio-curatorial staff will engage with both community members and the informatics staff for ongoing improvements. We have organized our objectives into the same structure as the Overall Component. These objectives are:

Aim 1: Provide a comprehensive model of biological knowledge focused on human biology

We will continue to support GO annotation efforts that provide the majority of curated GO annotations. We will capture knowledge from the literature using a more expressive model (LEGO, now called GO-CAM). We will improve our annotation efficiency workflow, not only because of the expressiveness of our shared annotation interface, but equally, because of the continued refinement, and conformance to, best practice guidelines we iteratively improve through biweekly case studies. At the same time, we will improve the biological content of the ontology. This synergistic improvement cycle between ontology development and annotation (biological model building) will increase efficiency on both sides.

Aim 2. Provide the “hub” for a broad community of scientists to extend and modify the GO

We will continue to expand and enhance our existing community annotation support. We will strengthen our collaborative partnerships with specialized groups who wish to provide GO annotations, from model organism focused groups, to domain interest groups such as for physiological systems (e.g. heart, kidney), to groups studying particular molecular pathways (e.g. Wnt signaling). We will improve the accessibility of the Common Annotation Framework user documentation. We will utilize a variety of communications channels to engage more research groups (e.g., case studies, videoconferences, training material, personal contacts). We will provide a dedicated contact person for coordinated annotation/ontology development efforts by external communities in specific areas of human biomedical interest. We will directly contact publishers and PubMed Central to explore how reciprocal contributions can help in our mission curate and computerize the literature.

Aim 3: Apply biological knowledge obtained in model organisms to human biology

We will generate inferential annotations from experimental results to a wide range of taxa using comparative evolutionary relationships. The focus will be on gene families with members in the human genome. Utilizing robust protein family trees, with persistent identifiers for ancestral nodes, GO biocurators will identify points at which functions first arose, and where these are lost, and propagate resulting annotations to all descendent extant proteins. We will complete a first pass of phylogenetic annotation of human genes. We will update phylogenetic annotations when the defining experimental annotations are updated.

Aim 4: Maintain the integrity and quality of new and existing annotations

We will identify and update annotations determined to be strong candidates for review, with high priority placed on review of human genes. We will recommend additional consistency checks for the GO informatics team to incorporate into the annotation quality assurance pipeline and to include as built-in within our annotation tools. We will organize annotation consistency reviews and provide comprehensive documentation and training.

Aim 5. Enable the scientific community to make full use of the GO data resources

We will provide open access to all of the annotations we produce via our Internet portal and Web services. We will provide dedicated user support, staffed by our biocurators and software developers. We will create and post video tutorials and run Webinars. We will document usage and parameterization for our GO enrichment analysis tool. We will provide the scientific community with computable biological models and we will continue to provide the standard GO files that are now so widely embedded in scientific workflows.