The Online Bioinformatics Resources Collection (OBRC)
Ansuman Chattopadhyay, Yi-Bu Chen, Jill Foust, Nancy H. Tannery and Barbara A. Epstein
Health Sciences Library System, University of Pittsburgh, 200 Scaife Hall, 3550 Terrace Street, Pittsburgh, PA 15261, USA.
Results – Figure 4. HSLS Molecular Biology Information Figure 8. Search OBRC using Vivísimo Clustering Engine (left); more Figure 9. Search OBRC using
Abstract Objectives Service Web Portal detailed view of the clustered topics (right) Zope®-based search engine
Contents and sources of OBRC
The primary objective of this project was to develop
In an attempt to provide easy access to the rapidly
growing number of Web-based resources for a new search strategy that would allow users to
bioinformatics and molecular biology research, we effectively search and locate the relevant online
have developed a searchable database, the Online resources for their specific information needs. The
Bioinformatics Resources Collection (OBRC) at the new strategy consists two major components:
University of Pittsburgh. The OBRC, currently
containing annotated links for 1780 peer reviewed 1. Create a centralized collection of information on major
free-access major online molecular databases and online biomedical databases and software tools.
software tools categorized in a hierarchical manner 2. Employ advanced search solutions, such as the Vivísimo
into 13 categories, was constructed using the HSLS Clustering Engine, to enhance the output of search results.
content management system built on the Zope Web
application server. We further implemented the
Vivísimo Clustering search engine, which Methods
automatically organizes the search results into
categories created dynamically based on the textual
information of the retrieved records. Instead of A. Source materials
browsing a long list of search results presented as
Figure 5. HSLS Online Bioinformatics Resources
a chronological list of hyperlinks, this clustering by 1. All validated databases published in NAR Annual
Collection at HSLS Web Site
topics helps to navigate and quickly identify the Database Issues from 2001 to 2007.
appropriate resources. As the largest online 2. All validated software tools published in NAR Annual
collection of its kind and the only one with Web Server Issues from 2004 to 2006.
advanced clustered search results, OBRC is aimed 3. Selected databases listed in the NAR Molecular Biology
at becoming a one-stop guided information solution Database Collection.
to the major bioinformatics and molecular biology 4. Selected databases and software published in other peer-
resources available on the Web. OBRC is available reviewed journals as well as other sources.
at the University of Pittsburgh's HSLS Web site
(http://www.hsls.pitt.edu/guides/genetics/obrc). B. Collection construction
The explosive growth of online
1. Information of each online resource was entered using the
HSLS content management system built on the Zope ®
Summary
bioinformatics resources Web application server.
2. For each entry, the information for the following fields are
We have successfully created the Online Bioinformatics Resources Collection,
The remarkable advance of genomic and proteomic entered: URL to the resource; name; links to PubMed
technology in the past decade has fueled an explosive abstract and other relevant documentation; brief covering the most widely used and authoritative bioinformatics databases and
growth of bioresearch data and literatures. description of the resource; highlights of the resource; software tools on the Web.
Hundreds of bioinformatics resources including index terms generated based on the content and scope of
databases and software tools in various fields of bio- the resource and domain expertise. The implementation of the Vivísimo Clustering Engine in OBRC enhances the
sciences are now publicly available on the Web. output of search results thus helping users to navigate through large number of
In the past decade, the number of online databases C. Vivísimo Clustering Engine implementation
listed in the Nucleic Acids Research Molecular Biology
results.
1. Queries are first processed by the Zope ® -based search
Database Collection has increased more than 14-fold
engine that leverages on Zope ® search tools. The rich content in OBRC coupled with the advance search features represents a
from 58 in 1996 to 858 in 2006 (Figure 1).
2. The results returned by the Zope ® based search engine novel search solution for online bioinformatics resources that will benefit
are then processed by the Vivísimo Clustering Engine on-
the-fly using the text information from the pre-determined
biomedical researchers at large.
fields.
3. The search results clustered by the Vivísimo Engine are
finally presented to the users. Future studies
Figure 6. OBRC sample record entry Figure 7. OBRC sample record display Further expand OBRC to include new databases and software published by NAR
as well as other journals.
Study user queries and user experiences to further improve the usability of
OBRC.
Explore new methods, such as constructing an embedded synonym mapping
protocol, to further improve the recall and precision.
The problems associated with searching growing Key References
number of online bioinformatics resources
Many researchers use Web search engines to locate a Chen YB, Chattopadhyay A, Bergen P, Gadd C, Tannery N. 2007. “The Online
database.
The information about the online resources is scattered in Bioinformatics Resources Collection at the University of Pittsburgh Health Sciences Library
various scientific journals. System--a one-stop gateway to online bioinformatics databases and software tools”. Nucleic
Few Web site currently provides a guided access point with Acids Research 35 (Database issue), D780-5.
searchable annotated links and to the majority of these
resources.
Searches using popular Web search engines, such as Nucleic Acids Research. Annual Database Issues and Annual Web Server Issues.
Google, are often ineffective mainly because:
Web search engines rank Web sites by popularity Calperin M.Y. 2006. “The molecular biology database collection: 2006 update” . Nucleic
rather than their relevance to researchers. Acids Research 34, D3-5.
Web search engines are unable to discriminate between
reliable and unreliable Web sites.
