Skip to content

This is a complete refactor of the original PHDeep, in preparation for publication

License

Notifications You must be signed in to change notification settings

1tilly/PHDeep2.0

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

9 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Current state

This code has not been tested since the refactoring. This is going to happen, when a large chunk of the code has been moved. Further, there are a few placeholder files/direcotires present, as the first step in the refactoring was the consideration of modules and components in this repository.

PHDeep - Pulmonary Hypertension Deep Learning pipeline

This project is a refactored version of my PhD project, based on the thesis "Deep Learning of regulatory sequence variation in Pulmonary Arterial Hypertension". It is still strongly under development.

Project structure

This project is modularised to be as adjustable as possible, while still (hopefully) easy to understand. Most module names should be self-explanatory. Something to point out are prediction and post_prediction. The former holding functions to actually run across either the reference genome or through vcf files, fetching the reference around the variants on the go. For this, the data_loading module is used to load the files. Post_prediction is for the handling of predicted epigenetic marks. Here, you will find the functions that compute the scores and make them accessible to the statistical/association testing.

Future work

These modules will be connected in Nextflow Pipelines, so that it is possible to run the pipeline from training-set generation over model training to prediction and post_prediction.

About

This is a complete refactor of the original PHDeep, in preparation for publication

Resources

License

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published