Integrating automated literature searches and text mining in biomarker discovery
2
ORAL PRESENTATION Open Access Integrating automated literature searches and text mining in biomarker discovery Maté Ongenaert * , Luc Dehaspe From Workshop on Advances in Bio Text Mining Ghent, Belgium. 10-11 May 2010 Background Epigenetics, and more specifically DNA methylation is a fast evolving research area. In almost every cancer type, each month new publications confirm the differentiated regulation of specific genes due to methylation and mention the discovery of novel methylation markers. The last decade, high-throughput methodologies are fre- quently used in the discovery of such methylation bio- markers. Examples of such analyses are re-expression experiments (using the demethylating agent 5-Aza-2′ - Deoxycytidine, followed by expression micro-array analysis); CpG microarrays such as the Illumina Human- Methylation27 BeadChip and large scale bisulfite sequencing. In order to evaluate and to prioritize possible methylation biomarkers, a literature search is a good starting point. However, manual searches are time- consuming (as hundreds of genes are to be searched, taking all their aliases into account) and the summari- zation of the found references is a real challenge. Therefore, it would be extremely useful to have an annotated, reviewed, sorted and summarized overview of all available data, published in methylation research in cancer. Results In a first stage, an automated literature retrieval and annotation tool was created, code-named GoldMine. This web-based application allows entering a list of genes, keywords and highlighting terms. Of the genes, all aliases are used to search PubMed abstracts, in com- bination with the keywords. The gene aliases, the key- words and the highlighting terms are highlighted in different colors as well as sentences with both a gene alias and a keyword. Abstracts are presented with decreasing scores that are assigned. Based on this framework, a cancer methylation data- base is created: PubMeth (as shown in Figure 1). Pub- Meth [1] is a cancer methylation database that contains genes that are reported to be methylated in various can- cer types. A query can be based either on genes (to check in which cancer types the genes are reported as being methylated) or on cancer types (which genes are reported to be methylated in the cancer (sub) types of interest). More recently, in the context of the SBO project on Functional Peptidomics, the MouseMining tool was developed to further exploit PubMeth results and com- parable literature summary data by combining them with experimental data. In a prototypical application, MouseMining was used to correlate statistics on the co- occurrence of anatomic categories and disease names to the expression profile of candidate biomarkers. Conclusions The generated methylation database in cancer is freely accessible at http://www.pubmeth.org. PubMeth is based on text mining of Medline/PubMed abstracts, combined with manual reading and annotation of preselected abstracts. The text mining approach results in increased speed and selectivity (as for instance many different aliases of a gene are searched at once), while the manual screening significantly raises the specificity and quality of the database. The summarized overview of the results is very useful in case more genes or cancer types are searched at the same time. Published: 6 October 2010 * Correspondence: [email protected] OncoMethylome Sciences, 4000 Liege, Belgium Ongenaert and Dehaspe BMC Bioinformatics 2010, 11(Suppl 5):O5 http://www.biomedcentral.com/1471-2105/11/S5/O5 © 2010 Ongenaert and Dehaspe; licensee BioMed Central Ltd.
-
Upload
mate-ongenaert -
Category
Documents
-
view
216 -
download
4
Transcript of Integrating automated literature searches and text mining in biomarker discovery
ORAL PRESENTATION Open Access
Integrating automated literature searches andtext mining in biomarker discoveryMaté Ongenaert*, Luc Dehaspe
From Workshop on Advances in Bio Text MiningGhent, Belgium. 10-11 May 2010
BackgroundEpigenetics, and more specifically DNA methylation is afast evolving research area. In almost every cancer type,each month new publications confirm the differentiatedregulation of specific genes due to methylation andmention the discovery of novel methylation markers.The last decade, high-throughput methodologies are fre-quently used in the discovery of such methylation bio-markers. Examples of such analyses are re-expressionexperiments (using the demethylating agent 5-Aza-2′-Deoxycytidine, followed by expression micro-arrayanalysis); CpG microarrays such as the Illumina Human-Methylation27 BeadChip and large scale bisulfitesequencing.In order to evaluate and to prioritize possible
methylation biomarkers, a literature search is a goodstarting point. However, manual searches are time-consuming (as hundreds of genes are to be searched,taking all their aliases into account) and the summari-zation of the found references is a real challenge.Therefore, it would be extremely useful to have anannotated, reviewed, sorted and summarized overviewof all available data, published in methylation researchin cancer.
ResultsIn a first stage, an automated literature retrieval andannotation tool was created, code-named GoldMine.This web-based application allows entering a list ofgenes, keywords and highlighting terms. Of the genes,all aliases are used to search PubMed abstracts, in com-bination with the keywords. The gene aliases, the key-words and the highlighting terms are highlighted indifferent colors as well as sentences with both a gene
alias and a keyword. Abstracts are presented withdecreasing scores that are assigned.Based on this framework, a cancer methylation data-
base is created: PubMeth (as shown in Figure 1). Pub-Meth [1] is a cancer methylation database that containsgenes that are reported to be methylated in various can-cer types. A query can be based either on genes (tocheck in which cancer types the genes are reported asbeing methylated) or on cancer types (which genes arereported to be methylated in the cancer (sub) types ofinterest).More recently, in the context of the SBO project on
Functional Peptidomics, the MouseMining tool wasdeveloped to further exploit PubMeth results and com-parable literature summary data by combining themwith experimental data. In a prototypical application,MouseMining was used to correlate statistics on the co-occurrence of anatomic categories and disease names tothe expression profile of candidate biomarkers.
ConclusionsThe generated methylation database in cancer is freelyaccessible at http://www.pubmeth.org. PubMeth is basedon text mining of Medline/PubMed abstracts, combinedwith manual reading and annotation of preselectedabstracts. The text mining approach results in increasedspeed and selectivity (as for instance many differentaliases of a gene are searched at once), while the manualscreening significantly raises the specificity and qualityof the database. The summarized overview of the resultsis very useful in case more genes or cancer types aresearched at the same time.
Published: 6 October 2010
* Correspondence: [email protected] Sciences, 4000 Liege, Belgium
Ongenaert and Dehaspe BMC Bioinformatics 2010, 11(Suppl 5):O5http://www.biomedcentral.com/1471-2105/11/S5/O5
© 2010 Ongenaert and Dehaspe; licensee BioMed Central Ltd.
Reference1. Ongenaert M, Van Neste L, De Meyer T, Menschaert G, Bekaert S, Van
Criekinge W: PubMeth: a cancer methylation database combining textmining and expert annotation. Nucleic Acids Res 2008, 36:D842-D846.
doi:10.1186/1471-2105-11-S5-O5Cite this article as: Ongenaert and Dehaspe: Integrating automatedliterature searches and text mining in biomarker discovery. BMCBioinformatics 2010 11(Suppl 5):O5.
Submit your next manuscript to BioMed Centraland take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at www.biomedcentral.com/submit
Figure 1 Scheme that illustrates the initial filling up of database using text mining. Aliases of genes and different keyword lists(methylation, cancer and detection-related) are highlighted in the abstract. At the same time, different parameters are counted and stored in aMySQL relational database. Afterwards, the data is ranked and manually reviewed.
Ongenaert and Dehaspe BMC Bioinformatics 2010, 11(Suppl 5):O5http://www.biomedcentral.com/1471-2105/11/S5/O5
Page 2 of 2
