Pathway curation: from isolated pathway knowledge to search-click-and-grow.
2012-10-31T15:40:21Z (GMT) by
<p>We here present an approach to address the problem of aggregating all knowledge supporting or contradicting biological pathways. With the increasing accuracy and high-throughput aspects of nowadays biological experimental research, the amount of data for known pathways is exploding. However, it is hard to match past knowledge on pathways against these new data and insights.</p> <p>Pathways, like signalling, metabolic, and transcription pathways, are abstractions of biological interactions. They provide a way to present large quantities of knowledge in a concise manner. The curation of pathways involves the integration of biological data from pluriform set of different knowledge source, such as databases, the literature or crowd mined expert knowledge. Current pathway approaches require large curation teams browsing through many websites. Whether it is PubMed, Uniprot, or any other relevant resource. Because most of the data is already in electronic form, data aggregation should be feasible. The different website layout are not helping, but many resources do come with a so-called application program interface (API), which allow access to the data. </p> <p>We demonstrate how existing pathway knowledge can be complemented with those further resources. </p> <p>For this, we use PathVisio (<a href="http://www.pathvisio.org/">http://www.pathvisio.org/</a>), which is an extensive desktop pathway editor and pathway analysis tool. As an analysis tool it allows the projection of experimental data on pathways. PathVisio shares a both an open source codebase and a native format (GPML) with WikiPathways (<a href="http://www.wikipathways.org/">http://www.wikipathways.org/</a>). WikiPathways is an open resource for biological signalling, pathway, and regulation pathways. It is working based on the same principles as Wikipedia where the community as a whole provides the knowledge. WikiPathways contains more than 2000 pathways covering 26 species. WikiPathways shares an open source codebase and its native format GPML with WikiPathways (http://www.wikipathways.org). Where WikiPathways requires a lean code base for optimized online performance, PathVisio being run as a desktop application, allows more complex algorithms and libraries to be included. This also enables the development of various plug-ins. <br>We have developed a plug-in called PathVisio Loom, which provides a framework for knowledge aggregation through menu guided additions. It allows the integration of biological knowledge available in the different formats (e.g. text mining data, interaction data available through webservices, Semantic Web Data or structure interaction data from local files). Starting from a single pathway object, after clicking the curator is given a set of known interactions aggregated from different online resources to select those relevant to the pathway under scrutiny. </p> <p><strong>With Pathvisio Loom, pathway knowledge integration transforms from a mainly manually editing process to a more click-and-grow process.</strong></p>