After the publication of this work , we became aware of several errors.
1. The headings of Table 1 and Table 2 have been swapped. Table 1 heading should read: “Summary of features and how they are calculated” while Table 2 heading should read: “Number of peaks found by Poisson thresholding of sliding window method.”
2. In Methods section, the last sentence of ‘Peak finding’ should read: “Finally, any peak windows separated by less than 50 (not 30) base pairs are merged into a single peak.”
3. In Results section, ‘Classification evaluation’, description of Figure 4B should read: “Recall is essentially the same as for the internal feature set, while a moderate improvement of precision is observed”. Description of Figure 4C should read: “Precision is better, relative to the internal feature set.”
4. One of the software packages we used, Genomic Ranges has been published recently, and can now be referenced .
The changes do not affect the correctness of the method described. We regret any inconvenience that these errors might have caused.
BMC Genomics 2013, 14(Suppl 5):S9. BioMed Central Full Text