GoekeLab / xpore

Identification of differential RNA modifications from nanopore direct RNA sequencing
https://xpore.readthedocs.io/
MIT License
131 stars 23 forks source link

Interpretation of results #151

Open zzf77 opened 1 year ago

zzf77 commented 1 year ago

Hi developer, I have successfully run my data with xpore. But I have a question, how could I distinguish which group has a higher m6A level in a single m6A motif, depending on the diff_mod_rate_con_vs_test? If I got a negitvae number in the diff_mod_rate_con_vs_test column, does it mean the m6A level of the control group at this site is lower than that of the test group? Thank you for your time! Best wishes.

yuukiiwa commented 1 year ago

Hi @zzf77,

To distinguish which group having a higher m6A level in a single m6A motif, you can look into the diff_mod_rate_<condition1>_vs_<condition2>, z_score_<condition1>_vs_<condition2>, and pval_<condition1>_vs_<condition2> columns.

The direction (either negative or positive) of diff_mod_rate_con_vs_test is alphabetically ordered. If you get a negative number in the diff_mod_rate_con_vs_test column, this means that test < con, which means the site is modified.

Thanks!

Best wishes, Yuk Kei

kwonej0617 commented 1 year ago

Hi, Yuk Kei. I have a question regarding this from your previous comment, 'If you get a negative number in the diff_mod_rate_con_vs_test column, this means that test < con, which means the site is modified.'.

Isn't diff_mod_rate_con_vs_test column what we would get from mod_rate_con (control, KO, unmodified) - mod_rate_test (WT, modified)? If so, if we get a negative number in the diff_mod_rate_con_vs_test, I think that mod_rate_test (WT, modified) should be greater than mod_rate_con (control, KO, unmodified). Also, when mod_rate_test (WT, modified) is grater than mod_rate_con (control, KO, unmodified), the site is likely to be modified. Could you please let me know if there's any misunderstanding?

Thank you so much!

zzf77 commented 1 year ago

Hi @zzf77,

To distinguish which group having a higher m6A level in a single m6A motif, you can look into the diff_mod_rate_<condition1>_vs_<condition2>, z_score_<condition1>_vs_<condition2>, and pval_<condition1>_vs_<condition2> columns.

The direction (either negative or positive) of diff_mod_rate_con_vs_test is alphabetically ordered. If you get a negative number in the diff_mod_rate_con_vs_test column, this means that test < con, which means the site is modified.

Thanks!

Best wishes, Yuk Kei

Hi, Yuk Kei. Thank you for your reply! If I got a negative number in the diff_mod_rate_con_vs_test column, this means that test < con, which means the site is modified. Does it means in this situation the m6A level of test < m6A level of con? If so, the positive number in the diff_mod_rate_con_vs_test column means the m6A level of test > m6A level of con?

Thanks! Best wishes.

vetmohit89 commented 9 months ago

Hello Yuk Kei,

can you please comment on it? I think @zzf77 has valid point.