Skip to content

This repository accompanies my research into the interpretability of DNA Damage Repair Outcome Predictors (DROPs). By analyzing these models using interpretability methods, we hope to uncover what features specifically are driving the accuracy of these prediction models.

Notifications You must be signed in to change notification settings

aaronwtr/interpreting-ml-based-drops

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Interpreting DNA Damage Repair Outcome Predictors (DROPs)

This repository contains all the code used and developed to implement DROPs and explain them with the help of the SHAP interpretability package.

The models that have been used in this research are FORECasT and Lindel, both logistic regression-based CRISPR/Cas9 induced DNA damage repair prediction models that predict the repair outcome distribution across different expected repair outcomes given a gRNA-target sequence.

The paper that resulted from this research can be found here.

About

This repository accompanies my research into the interpretability of DNA Damage Repair Outcome Predictors (DROPs). By analyzing these models using interpretability methods, we hope to uncover what features specifically are driving the accuracy of these prediction models.

Topics

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published