Large no. of reads go into "Nfail" column in pileup #583
Description
Dear authors,
I ran the modkit pileup command for modified bases a, h, and c. It appears that when I specify the 10th percentile threshold, I get an extremely low overall modification frequency (0.09%), since all mod probabilities are ~1. When I add a slightly conservative modification threshold, a large chunk of reads are going to the Nfail column.
I have added the log files for both runs of cytosine modifications.
Can you please advise as to what is the best way to move ahead in such a case?
Thank you!