Comparison with ENCODE. CHANCE can quickly compare the user's experiment with thousands of other experiments in the ENCODE repository. This figure shows CHANCE's output for H3K27me3 ENCODE data from the Broad Institute. On the x-axis is the IP over Input odds ratio that a randomly chosen read from the user's sample will lie in the union of all peaks of all ENCODE samples for the same transcription factor or epigenetic mark as the user's ChIP. The blue bell curve is a probabilistic model fitted to all available data. The blue circles denote all available data sets from ENCODE, and the red star is the user's sample. Intuitively, the user's sample has a poor overlap with the ENCODE data if the red star lies in the extreme left tail. The figure shows that the H3K27me3 sample is not an outlier when compared to other ENCODE samples.
Diaz et al. Genome Biology 2012 13:R98 doi:10.1186/gb-2012-13-10-r98