novoalab / EpiNano

Detection of RNA modifications from Oxford Nanopore direct RNA sequencing reads (Liu*, Begik* et al., Nature Comm 2019)
GNU General Public License v2.0
108 stars 31 forks source link

Which pre-trained model should be used in Epinano_Predict.py (RNA m6A and m5C) #84

Closed Stakaitis closed 3 years ago

Stakaitis commented 3 years ago

Hi,

1) In EpiNano/models folder are currently 7 models, which ones should be used for RNA m6A prediction if one does not have a WT vs KO setup and only wants to check 1 sample for modifications?

2) Is it possible to use Epinano for RNA m5C detection without WT vs KO setup?

Thank You! Rytis

Huanle commented 3 years ago

Hi Rytis,

In EpiNano/models folder are currently 7 models, which ones should be used for RNA m6A prediction if one does not have a WT vs KO setup and only wants to check 1 sample for modifications?

You can use all those trained without delta features.

Is it possible to use Epinano for RNA m5C detection without WT vs KO setup?

I have not tried predicting RNA m5C using SVM? But maybe you can give it a go by training some published data and use the trained models to make predictions. But without replicates, it will be hard to control for false positives.

Let me know if you need further information.

Huanle