The Informative Extremes: Using Both Nearest and Farthest Individuals Can Improve Relief Algorithms in the Domain of Human Genetics
A primary goal of human genetics is the discovery of genetic factors that influence individual susceptibility to common human diseases. This problem is difficult because common diseases are likely the result of joint failure of two or more interacting components instead of single component failures. Efficient algorithms that can detect interacting attributes are needed. The Relief family of machine learning algorithms, which use nearest neighbors to weight attributes, are a promising approach. Recently an improved Relief algorithm called Spatially Uniform ReliefF (SURF) has been developed that significantly increases the ability of these algorithms to detect interacting attributes. Here we introduce an algorithm called SURF* which uses distant instances along with the usual nearby ones to weight attributes. The weighting depends on whether the instances are are nearby or distant. We show this new algorithm significantly outperforms both ReliefF and SURF for genetic analysis in the presence of attribute interactions. We make SURF* freely available in the open source MDR software package. MDR is a cross-platform Java application which features a user friendly graphical interface.
Unable to display preview. Download preview PDF.
- 8.Williams, S.M., Canter, J.A., Crawford, D.C., Moore, J.H., Ritchie, M.D., Haines, J.L.: Problems with Genome-Wide association studies. Science 316(5833), 1841–1842 (2007)Google Scholar
- 13.Kira, K., Rendell, L.A.: A practical approach to feature selection, pp. 249–256 (1992)Google Scholar
- 14.Beretta, L., Cappiello, F., Moore, J.H., Barili, M., Greene, C.S., Scorza, R.: Ability of epistatic interactions of cytokine single-nucleotide polymorphisms to predict susceptibility to disease subsets in systemic sclerosis patients. Arthritis and Rheumatism 59(7), 974–983 (2008)CrossRefGoogle Scholar
- 15.The Wellcome Trust Case Control Consortium: Genome-wide association study of 14,000 cases of seven common diseases and 3,000 shared controls. Nature 447(7145), 661–678 (2007)Google Scholar
- 23.McKinney, B.A., Crowe, J.E., Guo, J., Tian, D.: Capturing the spectrum of interaction effects in genetic association studies by simulated evaporative cooling network analysis. PLoS Genet. 5(3), e1000432 (2009)Google Scholar
- 25.Kononenko, I.: Estimating attributes: Analysis and extensions of RELIEF. In: Bergadano, F., De Raedt, L. (eds.) ECML 1994. LNCS, vol. 784, pp. 171–182. Springer, Heidelberg (1994)Google Scholar
- 26.Sokal, R.R., Rohlf, F.J.: Biometry: the principles and practice of statistics in biological research, 3rd edn. W. H. Freeman and Co., New York (1995)Google Scholar
- 29.Shao, H., Burrage, L.C., Sinasac, D.S., Hill, A.E., Ernest, S.R., O’Brien, W., Courtland, H., Jepsen, K.J., Kirby, A., Kulbokas, E.J., Daly, M.J., Broman, K.W., Lander, E.S., Nadeau, J.H.: Genetic architecture of complex traits: Large phenotypic effects and pervasive epistasis. Proc. Nat. Acad. Sci. 105(50), 19910–19914 (2008)CrossRefGoogle Scholar
- 30.Robnik-Sikonja, M., Kononenko, I.: An adaptation of relief for attribute estimation in regression. In: ICML 1997: Proceedings of the Fourteenth International Conference on Machine Learning, pp. 296–304 (1997)Google Scholar