Here at Univ of Penn we have been running histone modification experiments on solexa and as a control we run input sample that has not been immunoprecipitated. In the controls, as well as the samples, we see a significant number, maybe a hundred, of short regions where tons of reads pile up. In the samples we see the same effects in the same regions and we assume they are not real because they also happen in the control. These regions tend to be consistent from sample to sample and run to run. This is AFTER strong repeat masking so should have nothing to do with repeats.
As a result we've had to run controls every time and remove those regions where this is happening. We are not sure if this is happening to everybody or if it's just us. Has anybody else had similar experience, or has anybody else gotten this to work without such artifacts? Any feedback is greatly appreciated, thank you.
As a result we've had to run controls every time and remove those regions where this is happening. We are not sure if this is happening to everybody or if it's just us. Has anybody else had similar experience, or has anybody else gotten this to work without such artifacts? Any feedback is greatly appreciated, thank you.
Comment