Additive Risk Survival Model with Microarray Data
Author Information
Author(s): Ma Shuangge, Huang Jian
Primary Institution: Yale University
Hypothesis
Can an additive risk model effectively identify influential genes and predict survival risks in high-dimensional microarray data?
Conclusion
The proposed additive risk model and Lasso approach can identify a small subset of genes with satisfactory prediction performance for lymphoma survival risks.
Supporting Evidence
- The study identified six probes with nonzero estimates that are biologically relevant to lymphoma.
- Chi-square tests showed significant differences in survival functions between risk groups based on gene expression.
- Five out of six identified probes had occurrence indices close to 1, indicating stability and importance.
Takeaway
This study looks at how certain genes can help predict how long patients with lymphoma might live, using a special math model to find the important genes.
Methodology
The study uses an additive risk model with Lasso for simultaneous estimation and gene selection based on microarray data.
Potential Biases
Potential biases may arise from the selection of genes and the assumptions of the additive risk model.
Limitations
The study's findings may not generalize to other types of cancer or larger datasets due to the small sample size.
Participant Demographics
The study included 101 untreated patients with mantle cell lymphoma, 92 of whom were classified as having MCL.
Statistical Information
P-Value
0.0003
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website