See Nkx2.5/star-GRCz11 for example.
Have to re-consider prePeak?
If no significant peak is detected, the current strategy produces a low threshold (3*standard deviation + median), which in turn gives us a bunch of garbage peaks. It would be nice to avoid this.
See
Nkx2.5/star-GRCz11for example.Have to re-consider
prePeak?If no significant peak is detected, the current strategy produces a low threshold (3*standard deviation + median), which in turn gives us a bunch of garbage peaks. It would be nice to avoid this.