 |
    |
Genetic Programming Neural Networks as a Bioinformatics Tool for Human Genetics
Marylyn D. Ritchie1, Christopher S. Coffey2, and Jason H. Moore1
1Center for Human Genetics Research, Department of Molecular Physiology and Biophysics, Vanderbilt University, 519 Light Hall, Nashville, TN 37232
ritchie@chgr.mc.vanderbilt.edu
moore@chgr.mc.vanderbilt.edu
2Department of Biostatistics, University of Alabama at Birmingham, Ryals Public Health Bldg., Rm. 327M, Birmingham, AL, 35294
CCoffey@ms.soph.uab.edu
Abstract. The identification of genes that influence the risk of common, complex diseases primarily through interactions with other genes and environmental factors remains a statistical and computational challenge in genetic epidemiology. This challenge is partly due to the limitations of parametric statistical methods for detecting genetic effects that are dependent solely or partially on interactions. We have previously introduced a genetic programming neural network (GPNN) as a method for optimizing the architecture of a neural network to improve the identification of gene combinations associated with disease risk. Previous empirical studies suggest GPNN has excellent power for identifying gene-gene interactions. The goal of this study was to compare the power of GPNN and stepwise logistic regression (SLR) for identifying gene-gene interactions. Using simulated data, we show that GPNN has higher power to identify gene-gene interactions than SLR. These results indicate that GPNN may be a useful pattern recognition approach for detecting gene-gene interactions.
LNCS 3102, p. 438 ff.
lncs@springer.de
© Springer-Verlag Berlin Heidelberg 2004