GoPubMed: exploring PubMed with the Gene Ontology

Abstract

The biomedical literature grows at a tremendous rate and PubMed comprises already over 15 000 000 abstracts. Finding relevant literature is an important and difficult problem. We introduce GoPubMed, a web server which allows users to explore PubMed search results with the Gene Ontology (GO), a hierarchically structured vocabulary for molecular biology. GoPubMed provides the following benefits: first, it gives an overview of the literature abstracts by categorizing abstracts according to the GO and thus allowing users to quickly navigate through the abstracts by category. Second, it automatically shows general ontology terms related to the original query, which often do not even appear directly in the abstract. Third, it enables users to verify its classification because GO terms are highlighted in the abstracts and as each term is labelled with an accuracy percentage. Fourth, exploring PubMed abstracts with GoPubMed is useful as it shows definitions of GO terms without the need for further look up. GoPubMed is online at www.gopubmed.org. Querying is currently limited to 100 papers per query.

Figure 1

User interface of GoPubMed. The screen-shot of GoPubMed displays the results for the query ‘levamisole inhibitor’ limited to 100 papers. On the left, part of the GO relevant to the query is shown and on the right the abstracts for a selected GO term. The search terms are highlighted online in orange and the GO terms in green. Right of each abstract is a list with all the GO terms for that abstract ordered by an accuracy percentage. For example, the term ‘P-glycoprotein’, which is a synonym for the GO term ‘xenobiotec transporting ATPase’, is found with 100% accuracy, while ‘lung development’ matches only with 72%, as only the word ‘lung’ occurs in the abstract. Synonyms, such as the term ‘P-glycoprotein’ above, are displayed in dark grey and the synonymous term is given in a tool-tip (please note that Mozilla-based browsers do not currently break lines in tool-tips). Moving the mouse over the term displays the definition of the term in a tool-tip. The ontology on the left shows the paths from the root of the ontology—‘cellular component’, ‘biological process’ and ‘molecular function’—to the currently selected GO term. The number in brackets behind each GO term in the ontology is the number of papers the GO term or any of its children occur in. In the figure, the path from ‘molecular function’ to ‘alkaline phosphatase’ is shown and the number 71 behind the term ‘alkaline phosphatase’ indicates that there are 71 papers mentioning alkaline phosphatase. Clicking on the term displays the relevant abstracts, which confirm that levamisole inhibits alkaline phosphatase. Overall, the number of papers containing a term and its children is a very good indicator to let users select the most frequent terms and, thus, best representatives. Instead of using the ontology to browse through abstracts, users can also display all the abstracts in the same order as in PubMed with the additional benefit of displaying the GO terms and search keywords. Users can also search within the ontology using the input field at the bottom of the ontology. GoPubMed searches are currently limited to 100 papers per query. Answering a query takes ∼20 s.

Figure 2

GoPubMed displaying the results for ‘levamisole inhibitor’ that contain the GO term phosphofructokinase.