Mining PubMed for Gene Relationships using Custom Ontologies

GeneCup is a tool to efficiently and comprehensively answer the question "What do we know about these genes and the topic I study?". GeneCup answers this question by searching PubMed to find sentences containing gene symbols and a custom list of keywords. These keywords are organized into a user defined ontology, which groups related keywords together for better organization of the results. Human GWAS findings from NHGRI-EBI GWAS catalog are also included in the search. These gene-keyword relationships are presented as an interactive graph and a table. As an example, we present a curated ontology for drug addiction with over 300 keywords organized into seven categories. Furthermore, deep learning is used to distinguish automatically the sentences describing systemic stress from those describing cellular stress. A user account (free) is needed to create custom ontologies and save the search results.

Example: Rgma Penk.

By default, the addiction ontology is used. Custom ontologies can be created in user accounts. Please choose keyword categories to be included in the search.
Up to 200 gene symbols can be entered in the box above. Because we do not search for relationships between genes, a list with more items can be split into multiple searches without affecting the results. Gene symbols can be separated by a space, a comma or a semicolon.