Skip to main content
Figure 7 | BMC Genomics

Figure 7

From: Improving the statistical detection of regulated genes from microarray data using intensity-based variance estimation

Figure 7

Implementation of outlier detection. (a) Measured (σM, gray) and pooled (σM'(Imin), black curve) standard deviation of M, plotted against Imin, with the outlier spots highlighted (black points), for Dataset 3. (Compare to Fig. 3b.) (b) Scatterplot of average condition X intensity vs. average condition Y intensity for Dataset 3, with p-values indicated in color. (c) Z-statistic using Imin pooling metric vs. "gold standard" t-statistic with outliers highlighted in black, for a 3-array subset and 20-array "gold-standard" subset of Dataset 4. Outlier Z-statistics calculated using the pooled standard deviation. (d) Z-statistic using Imin pooling metric vs. "gold standard" t-statistic, with outliers highlighted in black, for the same data in (c). Outlier Z-statistics calculated using the measured standard deviation, for Dataset 4 subset #2.

Back to article page