In a Nutshell - GoPubMed can answer questions!
Classical search engines retrieve documents GoPubMed answers questions! Consider the following questions and see below how GoPubMed helps answering them. Few clicks bring you from some general keywords to an answer.
Which diseases are associated with HIV?
Which are leading centers and scientists for liver transplanation?
Which biological process is the protein Rab5 involved in and where is it located
in the cell?
In which organisms is toluene degradation studied?
Which enzymes are inhibited by aspirin?
Which enzymes are important for congenital muscular dystrophy?
Which anatomical structure is affected by helicobacter pylori?
And here are the answers:
Question: Which diseases are associated with HIV?
Answer: Type "HIV" and wait for the tree on the left to appear. Go to "top five & more" and click on "disease". Among others hepatitis and tuberculosis are mentioned. Clicking on tuberculosis retrieves the relevant articles including statements such as "HIV and parasitic co-infections in tuberculosis patients".
Question: Which are leading centers and scientists for liver transplantation?
Answer: Type "liver transplantation" and click on statistics for term "liver transplantation" at the top of the results page. Among the top authors is "P Neuhaus" and among the cities is "Berlin". Prof. Peter Neuhaus works at the Charite Hospital in Berlin, Germany, is a leading specialist in the field.
Question: Which anatomical structure is affected by the
bacterium helicobacter pylori?
Answer: Type "helicobacter pylori", go to "top five & more" and open "anatomy" Among other terms "gastric mucosa" is listed. Hovering the mouse over the term reveals an explanation, which mentions that gastric mucosa is the lining of the stomach.
Question: Which biological process is the protein Rab5 involved in and where is it located in the cell?
Answer: Type "rab5" and wait for the tree on the left to appear. Go to "top five & more". Click on biological process shows "endcytosis" and clicking on "cellular component" shows "endosomes". Hovering over the terms displays brief explanations of what endocytosis and endosomes are.
Question: In which organisms is toluene degradation studied?
Answer: Type "toluene degradation" and wait
for the tree on the left to appear.
Go to "top five & more" and open "organisms". The
bacteria pseudomonas are listed first. A click retrieves the relevant articles
Question: Which enzymes are inhibited by aspirin?
Answer: Type "aspirin" and wait for the tree on the left to appear. Go to "hierarchy of content" and then "chemicals and drugs" and "enzymes and co-enzymes". From there always click the top child until you reach "cyclooxygenase 1" and "cyclooxygenase 2". Clicking reduces the articles to a few which mention that aspirin inhibits cyclooxygenases.
Question: Which enzymes are important for congenital muscular dystrophy?
Answer: Type "congenital muscular dystrophy" and wait for the tree on the left to appear. Go to "hierarchy of content" and then "chemicals and drugs", "enzymes and co-enzymes", "enzymes", "transferases". There are a number of articles with statements such as "glycosyltransferases has revealed a novel mechanism for congenital muscular dystrophy."
After performing a search, the resulting abstracts are annotated with your query keywords and GO and MeSH terms. The abstracts are grouped using the GO and MeSH terms, which appear in the text. Now the GO and MeSH hierarchies can be used to systematically explore the search results.
Note that only a subset of all GO and MeSH terms may be relevant to your query. This subset – more frequent terms - is presented on the left hand side. Sorting documents to a highly organised network facilitates the finding of relevant documents significantly.
The hierarchy of content shows the whole GO and MeSH ontologies. GO and MeSH serve as table of contents in order to structure the over 16 million articles of the MEDLINE data base.
GoPubMed is originaly based on this publication:
Andreas Doms and Michael Schroeder.
GoPubMed:
Exploring PubMed with the GeneOntology.
Nucleic Acid Research, 33(Web Server Issue):W783--W786, 2005
The GO provides a controlled vocabulary to describe gene and gene products in different organisms. GO is a knowledge network containing about 20.000 biological terms. It is built up as a directed acyclic graph starting from three basic areas namely
GO terms are classified into only one of the three branches of the ontology. Although the ontology is presented as a tree, it is a network with cross links. So it is possible to navigate to a term of interest on different paths. Hence, a term of interest can be reached from quite different points of view.
See geneontoly.org for more datails about Gene Ontology.
MeSH is the controlled vocabulary thesaurus from National Library of Medicine's. It consists of sets of terms in a hierarchical structure that permits searching at various levels of specificity. At the most general level of the hierarchy are very broad headings such as “Anatomy&rldquo; or “Diseases”. More specific headings are found at more narrow levels.
There are more than 110,000 MeSH concepts in GoPubMed. There are also thousands of cross-references that assist in finding the most appropriate MeSH concept. So it is possible to navigate to a term of interest on different paths. Hence, a term of interest can be reached from quite different points of view. From the eleven levels of the MeSH hierarchy, GoPubMed uses the parts:
Searching by concepts, authors, journals, etc is done by using the same search strategy as in PubMed. As examples of the more important PubMed search strategies can be mentioned: Title and Abstract [TIAB], Author [AU] and Affiliation [AD]. After the search is done the system shows how many citations PubMed has found for the query. The 1000 latest documents are displayed by GoPubMed. More citations can be retrieved using advanced search capabilities.
20 citations are shown in each page. Links to others pages using “go to page” and/or “Previous | Next” are available. In the abstracts found keywords for your query are highlighted in orange. GO and MeSH terms are highlighted in gray.
On the left hand side of the abstracts you can find the more frequent terms.
Clicking the +
symbol in front of the GO and MeSH terms opens a list of child terms
related to the parent one. Note that a GO and MeSH term can have more
than one parent. Thus it is possible to reach a term that represents
your topic of interest by navigating different paths through the
ontology trees. Above the tree the 5 most statistically relevant GO and
MeSH terms (frequent terms) related to your query are shown.
If your topic of interest is among them you can choose it directly.
Only the abstracts related to a selected GO or MeSH term are displayed.
The hierarchy of more frequent terms is a powerful instrument for the exploration of search results.
Try to explore the results concentrating on ontology terms, which are related to your field of interest.
For example for the query “parkinson [TIAB]” by exploring the frequent terms,
e.g. following Dopamine, the result can be reduced to 202 papers in just one mouse click.
Two very important utilities “Require” and “Exclude” are shown to the right of the selected term (in this example Dopamine). You can define filter rules mentioning or excluding one or more branches of the hierarchy.
“Which cellular component is the protein rab5 associated with? ”
Craig C. Mello and Andrew Z. Fire obtained the Nobel Prize in Physiology or Medicine in 2006.
“Do you know in which topics Craig C. Mello and Andrew Z. Fire are working on? ”
Craig C. Mello and Andrew Z. Fire obtained the Nobel Prize in Physiology or Medicine 2006 "for their discovery of RNA interference - gene silencing by double-stranded RNA"
“Which disease is rhodopsin involved in? ”
More than 4000 Diseases can be explored und related with your query.
“Which biological processes are inhibited by aspirin? ”
“Which organism is mostly involved in toluene degradation? ”
“Which proteins are most relevant for sickle cell anemia? ”
Sickle cell anemia defined as “a disease characterized by chronic hemolytic anemia, episodic painful crises, and pathologic involvement of many organs ” is mostly related to Blood Proteins, Hemeproteins and Membrane Proteins.
With GoPubMed 2500 Proteins can be related to your query.