Abstract
The discovery of biomarkers that are informative for cancer risk assessment, diagnosis, prognosis and treatment predictions is crucial. Recent advances in high-throughput genomics make it plausible to select biomarkers from the vast number of human genes in an unbiased manner. Yet, control of false discoveries is challenging given the large number of genes versus the relatively small number of patients in a typical cancer study. To ensure that most of the discoveries are true, we employ a knockoff procedure to control false discoveries. Our method is general and flexible, accommodating arbitrary covariate distributions, linear and nonlinear associations, and survival models. In simulations, our method compares favorably to the alternatives; its utility of identifying important genes in real clinical applications is demonstrated by the identification of seven genes associated with Breslow thickness in skin cutaneous melanoma patients.
Highlights
The discovery of biomarkers that are informative for cancer risk assessment, diagnosis, prognosis and treatment predictions is crucial
We propose several novel strategies based on the knockoff framework for variable selection subject to control for the false discovery rate
In our simulations, compared with the Difference in R-Squared (DRS) statistic, we found that the Risk Reduction in Boosting (RRB) statistic achieves better performance in terms of false discovery rate (FDR) control and of statistical power for variable selection (See Appendix C), with much lower computational burden
Summary
The discovery of biomarkers that are informative for cancer risk assessment, diagnosis, prognosis and treatment predictions is crucial. Many biomarkers have been proven to be very informative for clinical usage, with prominent examples such as BRCA1 and HER2 in breast cancer [1,2], EGFR in non-small-cell lung carcinoma [3] and PSA in prostate cancer [4]. Genes associated with disease-related clinical outcomes can be identified by linking a patient’s gene expression to the disease progression [5] or other disease phenotypes. By understanding the regulatory roles of these associated genes on various cancers, treatment strategies may be developed. For these reasons, many gene signatures have been discovered for a variety of cancers
Sign-in/Register to view highlights & summary 
Published Version (
Free)
Talk to us
Join us for a 30 min session where you can share your feedback and ask us any queries you have
Schedule a call