Regularized Gene Selection in Cancer Microarray Meta-Analysis
Author Information
Author(s): Ma Shuangge, Huang Jian
Primary Institution: Yale University
Hypothesis
The study proposes a new method for gene selection in cancer microarray meta-analysis that can effectively identify cancer-associated genes across multiple experiments.
Conclusion
The MTGDR provides an effective way of analyzing multiple cancer microarray studies and selecting reliable cancer-associated genes.
Supporting Evidence
- The MTGDR method outperformed traditional methods in identifying true positive genes.
- Simulation studies showed that MTGDR can effectively select genes with joint effects on cancer.
- Analysis of pancreatic and liver cancer studies confirmed the effectiveness of the MTGDR method.
Takeaway
This study created a new method to find important genes in cancer by looking at many experiments together instead of just one at a time.
Methodology
The study uses a new approach called Meta Threshold Gradient Descent Regularization (MTGDR) for gene selection in cancer microarray data.
Limitations
Some genes showed inconsistent signs across studies, and detailed information on several identified genes was not available.
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website