Identifying Key Genes in Triple-Negative Breast Cancer Using Machine Learning
Author Information
Author(s): Ghazal Hany, El-Absawy El-Sayed, Ead Waleed, Hasan Mohamed E.
Primary Institution: China Medical University
Hypothesis
Can machine learning improve the identification of biomarkers in triple-negative breast cancer?
Conclusion
Machine learning-based gene selection significantly enhances the identification of hub genes that can serve as biomarkers for triple-negative breast cancer.
Supporting Evidence
- 27 genes were selected as differentially expressed using machine learning techniques.
- Hub genes identified include ESR1, FOXA1, GATA3, XBP1, GREB1, AR, and AGR2.
- Machine learning models showed high accuracy in distinguishing between TNBC and non-TNBC.
Takeaway
Researchers used computers to find important genes in a type of breast cancer called triple-negative breast cancer, which can help doctors choose better treatments.
Methodology
The study utilized data from GEO and TCGA to identify differentially expressed genes using LIMMA and edgeR algorithms, followed by machine learning feature selection.
Potential Biases
Potential biases may arise from the limited sample diversity and the reliance on existing datasets.
Limitations
The study is limited by the size of the datasets and the inherent heterogeneity of triple-negative breast cancer.
Participant Demographics
The study included samples from both TNBC and non-TNBC patients, with a focus on young women and women of African or Hispanic descent.
Statistical Information
P-Value
p<0.01
Statistical Significance
p<0.01
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website