Mining Cancer Biomarkers from Gene Expression Data
Author Information
Author(s): Fabien Campagne, Lucy Skrabanek
Primary Institution: Weill Medical College of Cornell University
Hypothesis
Can expressed sequence tags (ESTs) be mined to identify cancer biomarkers?
Conclusion
The study identifies a set of 200 cancer biomarkers (HM200) that are predictive across various tumor types and classification tasks.
Supporting Evidence
- HM200 markers achieved the best or second best classification performance in 79% of the evaluations.
- The approach generated less than 22% false discoveries when applied to combined human and mouse whole genome screens.
- HM200 genes were found to be enriched in oncogenes and tumor suppressor genes.
- Microarray validation confirmed that HM200 genes are effective predictors in cancer classification tasks.
Takeaway
Researchers found a group of genes that can help identify different types of cancer, which could lead to better tests for diagnosing cancer early.
Methodology
The study used expressed sequence tags to identify differentially expressed genes in cancer and validated findings through microarray data analysis.
Potential Biases
Potential bias from the selection of gene lists and the inherent variability in gene expression data.
Limitations
The study may have biases due to the varying proportions of EST libraries from tumor and non-tumor tissues.
Statistical Information
P-Value
0.0011
Statistical Significance
p<0.001
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website