Software for
Regularized ROC for Disease Classification and Biomarker Selection
(Last Update: Dec. 18th, 2005. By Shuangge Ma)
Shuangge Ma, Ph.D.
Collaborative Health Studies Coordinating Center and Department of
Biostatistics
University of Washington, Seattle, WA 98195
Email: shuangge@u.washington.edu
Jian Huang, Ph.D.
Department of Statistics & Actuarial Science and
Program in Public Health Genetics
University of Iowa, Iowa, IA 52242
Email: jian@stat.uiowa.edu
Introduction:
We provide the software (written in R) for estimation and variable
selection using the regularized ROC approach proposed in
the paper
Regularized ROC method for disease classification and biomarker selection
with microarray data published in Bioinformatics . The
paper is coauthored by Shuangge Ma and Jian Huang.
Below is the abstract of the paper. We strongly recommand you reading the
full paper before using the software.
Abstract
An important application of microarrays is to discover genomic
biomarkers, among tens of thousands of genes assayed, for disease
classification. Thus there is a need for developing statistical
methods that can efficiently use such high-throughput genomic
data, select biomarkers with discriminant power and construct
classification rules.
The ROC (receiving operator characteristic) technique has been
widely used in disease classification with low dimensional
biomarkers because (1). it does not assume a parametric form of
the class probability as required for example in the logistic regression
method; (2). it accommodates case-control designs; and (3). it
allows treating false positives and false negatives differently.
However, due to computational difficulties, the ROC based classification
has not been used with microarray data. Moreover,
the standard ROC technique does not incorporate built-in biomarker
selection.
We propose a novel method for biomarker selection and
classification using the ROC technique for microarray data. The
proposed method uses a sigmoid approximation to the area under the
ROC curve as the objective function for classification and the
threshold gradient descent regularization method for estimation
and biomarker selection. Tuning parameter selection based on the
V-fold cross validation and predictive performance evaluation are also
investigated. The proposed approach is demonstrated with
a simulation study, the Colon
data and the Estrogen data. The proposed
approach yields parsimonious models with excellent classification
performance.
Special Notes:
We would like to thank you for your interest in our study. You are
encouraged to try out the software. However, please note:
Please kindly inform us, if you see any error of the software.
Documentation:
Source Code and a Toy Example:
The following files are the source code and a toy example
demonstrating the use of the software. The software is written using R and
has been tested under standard Linux operating system. Certain
modifications may be needed for different operating systems.
The data generation code can be downloaded here.
Sample outcome data.
Sample covariate data.
The cross validation code generates two files. The first one shows the CV score as a function of iterations. The second one identifies the number of iterations needed from the first file.
Note that we randomly permute the data in the cross validation to avoid any "hidden pattern" (for example the first half subjects all from the case group). So when running the cross validation code, you may get quite similar, however different results.
For cross validation over different threshold, you need to change it manually. Adjustment of the "increase" value may be needed.
Acknowledgment:
The work of Ma is partially supported by the NIH grant N01-HC-95159. The work of Huang is supported in part by the NIH grant HL72288.
For
questions, comments or suggestions, email shuangge@u.washington.edu
or jian@stat.uiowa.edu.