Set up a benchmark similar to the SEACR paper (Fig 3a) that iteratively resamples peaks and tests what proportion of them are "correct". The "correctness" can be assessed, for example, by seeing how many of our peaks are in ENCODE.
Repeat these benchmarks separately for each combination of:
dataset (antibodies, concentrations, histone marks) and - peak calling method (MACS2, SEACR, SEACR with different thresholds)
Set up a benchmark similar to the SEACR paper (Fig 3a) that iteratively resamples peaks and tests what proportion of them are "correct". The "correctness" can be assessed, for example, by seeing how many of our peaks are in ENCODE.
Repeat these benchmarks separately for each combination of: