permGPU-package: packageTitlepermGPU
In permGPU: Using GPUs in Statistical Genomics

Description Details Note Author(s) References

This package can be used to carry out permutation resampling inference using GPUs, as described in I. D. Shterev, S. H. Jung, S. L. George and K. Owzar. "permGPU: Using Graphics Processing Units in RNA Microarray Association Studies", BMC Bioinformatics 11(1), 2010. Currently the package supports six test statistics: the t and Wilcoxon tests, for two-sample problems, the Pearson and Spearman statistics, for non-censored continuous outcomes, and the Cox score and rank score test (Jung et al, 2005), for right-censored time-to-event outcomes. In addition to the test statistics and the corresponding marginal permutation P-values, the package produces family-wise error adjusted P-values using a single-step procedure (Westfall and Young, 1993).

The DESCRIPTION file: packageDESCRIPTIONpermGPU packageIndicespermGPU

To build this package, the CUDA SDK (version 2.3 or higher) must be installed on the system. Specifically, the nvcc compiler must be in the path and the CUDA_HOME must be properly defined. For example, if the SDK kit is installed under /usr/local/cuda then the CUDA_HOME variable needs to be set to /usr/local/cuda . The SDK can be obtained from https://www.nvidia.com The CUDA_HOME variable can also be explicitly defined in permGPU/src/Makefile. The maximum number of patients for the current version is 1000.

The R environment variables R_LIB and R_INCLUDE need to be correctly configured to build the package from source. Alternatively, these can be set in permGPU/src/Makefile.

To build this package, a number of C++ classes and functions for random number generation (available from https://www.agner.org/random/ under a GPL license) and a C++ template for calculating ranks (available from https://sites.google.com/site/jivsoft/Home/compute-ranks-of-elements-in-a-c—array-or-vector under a BSD license) are needed. The requisite files are included in the package source code tar ball. In future releases, these functionalities will be replaced by native R functions from R.h and Rmath.h.

I. D. Shterev, S.-H. Jung, S. L. George and K. Owzar

Maintainer: I. D. Shterev <i.shterev@duke.edu>

Shterev, I.D., Jung, S.-H., George S.L., Owzar K. permGPU: Using graphics processing units in RNA microarray association studies. BMC Bioinformatics 2010, 11:329.

Jung, S.-H., Owzar K., George, S.L. (2005). A multiple testing procedure to associate gene expression levels with survival. Statistics in Medicine. 24(20), 3077–88.

Westfall, P.H. and Young, S.S. (1993). Resampling-Based Multiple Testing: Examples and Methods for P-value Adjustment, Wiley-Interscience, New York.

permGPU documentation built on Feb. 10, 2021, 5:07 p.m.