Extracting consistent knowledge from highly inconsistent cancer gene data sources

BMC Bioinformatics

Research Article

Jing Wang¹  Lin Zhang¹  Jing Zhu¹  Yuannv Zhang²  Wenyuan Zhao²  Xue Gong²  Lixin Cheng²  Yunyan Gu²  Ruihong Wu²  Zheng Guo³ 

[1] Bioinformatics Centre, School of Life Science, University of Electronic Science and Technology of China, 610054, Chengdu, China;College of Bioinformatics Science and Technology, Harbin Medical University, 150086, Harbin, China;College of Bioinformatics Science and Technology, Harbin Medical University, 150086, Harbin, China;Bioinformatics Centre, School of Life Science, University of Electronic Science and Technology of China, 610054, Chengdu, China;

关键词: Gene Ontology;    Gene List;    Cancer Gene;    Random Chance;    Cancer Gene Census;   

DOI  :  10.1186/1471-2105-11-76

 received in 2009-09-15, accepted in 2010-02-05,  发布年份 2010

来源: Springer

【 摘 要 】

BackgroundHundreds of genes that are causally implicated in oncogenesis have been found and collected in various databases. For efficient application of these abundant but diverse data sources, it is of fundamental importance to evaluate their consistency.ResultsFirst, we showed that the lists of cancer genes from some major data sources were highly inconsistent in terms of overlapping genes. In particular, most cancer genes accumulated in previous small-scale studies could not be rediscovered in current high-throughput genome screening studies. Then, based on a metric proposed in this study, we showed that most cancer gene lists from different data sources were highly functionally consistent. Finally, we extracted functionally consistent cancer genes from various data sources and collected them in our database F-Census.ConclusionsAlthough they have very low gene overlapping, most cancer gene data sources are highly consistent at the functional level, which indicates that they can separately capture partial genes in a few key pathways associated with cancer. Our results suggest that the sample sizes currently used for cancer studies might be inadequate for consistently capturing individual cancer genes, but could be sufficient for finding a number of cancer genes that could represent functionally most cancer genes. The F-Census database provides biologists with a useful tool for browsing and extracting functionally consistent cancer genes from various data sources.

【 授权许可】

Unknown   
© Gong et al; licensee BioMed Central Ltd. 2010. This article is published under license to BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.