Do Two Machine-Learning Based Prognostic Signatures for Breast Cancer Capture the Same Biological Processes?
Author Information
Author(s): Yotam Drier, Eytan Domany
Primary Institution: Weizmann Institute of Science
Hypothesis
Do different prognostic gene lists for breast cancer capture similar underlying biological processes?
Conclusion
The study found that the only biological process significantly represented by both prognostic signatures is cell proliferation.
Supporting Evidence
- The only biological process significantly represented by both signatures is cell proliferation.
- Claims of wider concordance between the biological processes captured by the two prognostic signatures were found lacking statistical rigor.
- Enrichment analysis was performed while controlling for multiple testing.
Takeaway
The study looked at two gene lists used to predict breast cancer outcomes and found they mostly agree on one thing: they both relate to how fast cells grow.
Methodology
The study performed enrichment analysis on two established prognostic gene lists while controlling for multiple testing.
Potential Biases
Potential biases may arise from the selection of gene lists and the methods used for enrichment analysis.
Limitations
The analysis may not capture all relevant biological processes due to the focus on specific gene lists.
Statistical Information
P-Value
0.015
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website