search

GoekeLab / xpore

Identification of differential RNA modifications from nanopore direct RNA sequencing
https://xpore.readthedocs.io/
MIT License
136 stars 22 forks source link

Read-level modifications #26

Open pabloacera opened 4 years ago

pabloacera commented 4 years ago

Hi @ploy-np thanks for the tool, it works really well!

In the output:

id,position,kmer,diff_mod_rate_KO_vs_WT,pval_KO_vs_WT,z_score_KO_vs_WT,mod_rate_KO-rep1,mod_rate_WT-rep1,coverage_KO-rep1,coverage_WT-rep1,mu_unmod,mu_mod,sigma2_unmod,sigma2_mod,conf_mu_unmod,conf_mu_mod,m
ENST00000273480.3,355,CTGTG,-0.08347772875845261,0.00045545433099164034,-3.505673042879497,0.9165081871312406,0.9999859158896932,135.00000000000003,71.0,99.48265403939226,104.17836035668046,28.4052619532913
ENST00000273480.3,593,CGATG,0.3427671649567362,7.343327455223234e-07,4.952056501330644,0.7507215520963253,0.4079543871395891,145.0,69.0,117.18091740473061,124.27790948192207,30.599099884885668,8.92729375396

If i'm not wrong, every line gives information per site right? I was wondering how can I get read-level modification information, like a probability or so, Is that possible or will be included in the future? Thanks

ploy-np commented 4 years ago

Hi @pabloacera,

That's good to hear :)

Yes, you can get the read-level information by using the --save_models option when you run xpore-diffmod. You will get all the learned parameters stored in a .hdf5 file for each transcript/gene. However, I haven't provided a script to extract those information in this version. Anyway, you can explore it yourself. The read-level parameter is called 'z', and you can find more details regarding the model parameters in our preprint. Hope this can help for the moment!