Chris Evelo, Maastricht University




To really get ahead with complex health problems like cancer and diabetes we need to become better at combining different types of studies, including large scale genomics and genetics studies and we need to learn to better combine such studies with existing biological knowledge. Typically this leads to questions like “I did this study with high-fat low fat diet comparison in mice and looked at the transcriptomics results in liver, fat and muscle. Has anyone else perform a study like thus and publish the data, maybe for proteomics? Can I find that in one of these open data repositories?”. Or, “I did that, can I find which biological pathways are affected most and whether any of the proteins in that pathway is a known target for an existing drug?”.  Or even “I did that study, could I find another study that yielded the same kind of biological results even if it was from a different research field with a completely different result?”. To answer these kind of questions we need to describe studies and study results, structure knowledge allow mapping of “equal” things with different identifier schemes and essentially do a lot of mapping to and between ontologies. More and more of that is getting real. I will discuss how ISA-tab inspired our study capture tool in which we use templates to get consistent ontology choices. How we use identifier mapping to allow analyses and visualisation of different kinds of data of "domain" annotated knowledge. And finally, how we use all kinds of mappings to extend knowledge captured in pathways with other information in databases. In that way I will hop, step and jump through ISA-tab, the phenotype database, WikiPathways and Open PHACTS using ontologies and mappings as a jumping pole.



Chris Evelo is the head of the department of Bioinformatics at Maastricht University. He is involved in a number of initiatives relevant to this topic. The phenotypic database initiative that uses ISA-tab based approaches to describe studies, samples and assays and links to data processing pipelines and actual data at different stages of processing for many technologies (see WikiPathways and its accompanying tools to structure existing knowledge in biological pathways and make it available for usage in analytical tools and Open PHACTS which builds a large Open Pharmacological Knowledge space based on semantic web technology for drug discovery.