search

gaoyubang / nanom6A

MIT License
18 stars 11 forks source link

about validation set used in your paper #19

Open xieyy46 opened 2 years ago

xieyy46 commented 2 years ago

Hi nanom6a team! I noticed that you used m6A sites identified by miCLIP from public data (Nanopore direct RNA sequencing maps the complexity of Arabidopsis mRNA processing and m6A modification) in Fig.3a of your paper, and the number of m6A sites identified by miCLIP is 4685, however in paper Nanopore direct RNA sequencing maps the complexity of Arabidopsis mRNA processing and m6A modification that numbr is 93046. I'd want to know how you got 4685 m6A sites from 93046 site? image

gaoyubang commented 2 years ago

I fillter m6A sites with RRACH motif

image

xieyy46 commented 2 years ago

Hi yubang! Your script (seq=loader.fetch(chro,pos-5,pos+5)) meant that you will save sites if the flanking 5bp (11bp in total) contains RRACH, though the flanking 2bp (5bp in total) may not == RRACH?