Using Genetic Programming to Identify Nodal Status in Bladder Cancer
Author Information
Author(s): Mitra Anirban P, Almal Arpit A, George Ben, Fry David W, Lenehan Peter F, Pagliarulo Vincenzo, Cote Richard J, Datar Ram H, Worzel William P
Primary Institution: University of Southern California Keck School of Medicine
Hypothesis
Can genetic programming analyze gene expression profiles to accurately determine nodal status in bladder cancer?
Conclusion
The study shows that genetic programming can effectively classify nodal status in bladder cancer using gene expression data.
Supporting Evidence
- The classifier rules demonstrated 81% accuracy on the validation set.
- Using only three genes, the accuracy reached 90% for nodal positivity.
- ICAM1, MAP2K6, and KDR were identified as key genes in nodal metastasis.
- Statistical analysis showed significant gene usage probabilities.
- Cross-validation techniques were employed to ensure robustness of the classifier.
Takeaway
Researchers used a computer program to look at genes in bladder cancer samples to see if they could tell if the cancer had spread to the lymph nodes.
Methodology
The study analyzed gene expression from 60 bladder cancer patients and 5 normal controls using genetic programming to classify nodal status.
Potential Biases
Potential bias in gene selection due to unequal distribution of nodal status across tumor stages.
Limitations
The study had a small sample size and may not generalize to larger populations.
Participant Demographics
60 patients with bladder cancer and 5 normal controls, with varying tumor stages.
Statistical Information
P-Value
p<0.00001
Statistical Significance
p<0.05
Digital Object Identifier (DOI)
Want to read the original?
Access the complete publication on the publisher's website