Skip to main content
Figure 2 | BMC Bioinformatics

Figure 2

From: Filtering, FDR and power

Figure 2

Filter statistic distribution for all features (blue histogram) and separately for features for which H 0 holds (blue line) and H a holds (red line) for one simulated dataset. Each one of these filter statistics leaves features out with small statistic values. The vertical gray lines mark deciles of the distribution for all features, so that if 10% of the features must be left out, then they are the ones with value of the filter statistic to the left of the first vertical gray line. The last vertical line leaves 1% of the data.

Back to article page