Skip to main content
Figure 1 | Genome Biology

Figure 1

From: Chipper: discovering transcription-factor targets from chromatin immunoprecipitation microarrays using variance stabilization

Figure 1

Three replicate two-channel Chip2 experiments performed on Sko1 [22] were variance-stabilized. (a-c) Distributions of the Δh values obtained. Shaded gray areas indicate kernel-smoothed densities estimated from data. Magenta curves estimate the distribution of scores expected of unbound intergenic regions (IGRs) by fitting a normal distribution to the negative Δh side of the distribution. Sufficient statistics (mean, variance) of each fitted distribution are used to standardize the Δh distributions to a score z i for each replicate. (d) The distribution of the average score over all three replicates. We computed a p value for each IGR under the null hypothesis that it is unbound, using the curve fitted to the negative portion of the empirical distribution.

Back to article page