search

darelab2014 / Dinopore

DInoPORE: Direct detection of INOsines in native RNA with nanoPORE sequencing
16 stars 3 forks source link

classref file #3

Closed ssscj closed 1 year ago

ssscj commented 1 year ago

Hi, Thanks for developing Dinopore. I want to use Dinopore to quantify the A-to-I editing events in my nanopore data, so I followed the steps to process the data. In step 5 and step 6 of testing path, classref file is required, and I don't know what this classref file is . In the training path step5, note 2 says classref file is the groundtruth_training_file. I want to use Dinopore to predict the edit rate of my data, so I don't have the ground truth file. How could I use Dinopore? Thanks.

Best regards, Chujie

skchronicles commented 1 year ago

@ssscj Is there any update on this issue? I am also in the same situation. I am not sure what I should be providing as the ground truth file. I am not sure if this file is only being used to assess accuracy/precision of the model, or if it is being used for something else more important.

@darelab2014 @hengjwj @tramanh183 If you could provide some clarification, that would be awesome!

Thank you again for your time and help. I appreciate it!

hengjwj commented 1 year ago

Hi @ssscj , @skchronicles,

Sorry for the late reply and thanks for trying out our method. The classref file is used to assess the model's prediction based on Illumina sequencing data. The file for training and testing are of different formats. I've moved the files over from Code Ocean for convenience. For the training file, please see /Dinopore/code/misc/Example_ground_truth_training.txt. For testing, please see /Dinopore/code/misc/Example_classref.txt.