Concept Explorer (CE)
The Concept Explorer (CE) targets clinicians and other scientists involved in designing and fine tuning clinical trials (CTs).
What current need does the Concept Explorer address?
CT research faces a number of challenges including formulating relevant research hypotheses, protocol design, appropriate inclusion/exclusion criteria, end point selection and others. As the need for therapy differentiation, more and better evidence, and CT cost optimization increases, so the need for better-designed CTs increases too. Such a benefit is expected to result from an increased ability to address the challenges in CT research. This in turn means a better understanding of the underlying biology, design pathophysiology and drug mechanism of action.
The Concept Explorer supports this requirement by providing a tool that helps scientists and clinicians uncover non-obvious, biologically relevant correlations through big data analytic approaches of the scientific literature.
How Does the Concept Explorer work?
The CE supports what it calls “browse directed search” (BDS). In BDS you start with a “concept of interest”, say a specific drug such as “tysabri”, and explore correlations with other concepts, say biological targets that the drug may be hitting. To perform this exploration, the user would specify in CE the name of the drug (e.g. “tysabri”) and then select the “concept class” she is interested in, say “targets”. The CE would then show a network of the targets that are linked to tysabri. Some of these will be known to the researcher and (hopefully) some will be novel to her.
Selecting ones that look interesting the researcher may then continue her exploration by browsing additional correlations and by drilling down to access the bibliography (one or more supporting articles) from which the correlation was extracted. In this manner, the researcher can formulate new hypotheses for information on the mechanism of action of a drug, or possible side effects that would inform patient exclusion criteria, protocol end points or other parameters relevant to the design of the clinical trial.
With the insights gained from the CE, the clinician can then proceed to define and manage the clinical trial using the other tools available via the p-medicine platform.
Concept Explorer Features
- Comprehensive coverage of Medline
- Continuous update
- Biannual update of terminology (ontology of genes, diseases, pathways, drugs etc.) – over 55 million terms monitored
- Ontologies used internally: MESH, GO, UMLS, REACTOME
- Covers concept classes: Gene, Pathway, Disease, Drug, Cell line, Biosystems, Organism, Infectious Organism, PTM
- Displays concept correlations as a graph, where nodes represent concepts (such as the drug “tysabri” or the pathway “apoptosis”) and links denote correlations between the two connected nodes (i.e. that the two nodes co-occur in one or more publications)
- Interface supports the following commands:
- Explore: specify the name of a concept of interest (e.g. a drug name) and a concept class of interest (e.g. “genes”). CE displays a graph showing the concept of interest and related concepts belonging to the specific class.
- Connect: finds and shows (if they exist) correlations between a selected concept from the graph
- Bridge: the user selects two concepts from the graph (e.g. two pathways) and specifies a concept class of interest (e.g. “genes”). The CE displays any gene that is correlated to both the selected concepts (in the example pathways)
- Specify Return nodes: The CE displays a specific number of related concepts, starting with the more strongly correlated ones
- Toggle bibliography: avoids display clutter by allowing the use to toggle the display of the underlying bibliography on and off
- Free Use: sign in with p-medicine portal account
- Runs on modern browsers
- Available on PCs and tablets
- Based on HTML 5 technology
- REST API
Contact Biovista at http://www.biovista.com/about/contact/