@article{Chen_Su_2019, title={Overcoming Adverse Effects of Correlations in Microarray Data Analysis}, volume={11}, url={http://rgnpublications.com/journals/index.php/jims/article/view/1214}, DOI={10.26713/jims.v11i2.1214}, abstractNote={<p>Due to the existence of the strong correlation between microarray gene expression levels, procedures which are commonly used to select the significant genes between two or more phenotypes cannot overcome the main problems: high instability of the number of false discoveries and low power. It may be impossible to completely understand these correlations due to the complexity of the biology nature. Gordon et al. [1] proposed a new multiple testing procedure to balance type <em>I</em> and <em>II</em> errors in an optimal way. However, the correlation structure of microarray data is still the main obstacle standing in the way of various gene selection procedures. To remove this obstacle, we improved the statistical methodology by exploiting the properties associated with the low dependency of the so-called delta-sequence proposed in Klebanov et al. [4]. Our study showed a similar behavior has been observed that both the mean and the standard deviation of the number of false positives are monotonically decreasing as a function of the threshold parameter. In addition, working with pairs, we have substantial reduction in both numbers, which means we gain power and stability in our new study.</p>}, number={2}, journal={Journal of Informatics and Mathematical Sciences}, author={Chen, Linlin and Su, Haiyan}, year={2019}, month={Jul.}, pages={209–219} }