search

gaoyubang / nanom6A

MIT License
19 stars 11 forks source link

minimam coverage when outputting results #13

Closed PengNi closed 3 years ago

PengNi commented 3 years ago

Hi @gaoyubang ,

I noticed that in nanom6A, only sites which are supported with at least --support number of reads can be saved into the results. However, I don't quite understand the code in predict_sites.py L#L405:

if nummod<numlimit:
    continue

Was it on purpose that nummod was used in here? Or it should be numtotal?

Thanks very much!

Best, Peng

gaoyubang commented 3 years ago

What we want is the number of reads supported at this modification site

PengNi commented 3 years ago

@gaoyubang , thanks for your reply. So on what purpose you set like this? Does you want to output only the sites with high methylation frequency, or only the sites with sufficient coverage of reads? Or because of some other biological reasons?

gaoyubang commented 3 years ago

we want to output m6A sites with high methylation level

PengNi commented 3 years ago

we want to output m6A sites with high methylation level

@gaoyubang , thanks, I understand. Though in my opinion, for this purpose, maybe using numtotal and fre together would be more proper, cause there are cases like, for example, a site has 20 methylated reads and 100 unmethylated reads, mathematically.

Best, Peng