Skip to content

A modified version of Deep learning MLP model to rank pegRNA sequence

Notifications You must be signed in to change notification settings

Ditean/pegRNA_rank

 
 

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 

History

26 Commits
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

Modified atgRNA_rank

This is a modified version of the deep learning MLP model to rank atgRNA sequence by the Abudayyeh-Gootenberg Lab. Please refer to and cite their paper for information about atgRNA. This fork is more intended for personal use rather than development for general use.

Yarnall, M. T., Ioannidi, E. I., Schmitt-Ulms, C., Krajeski, R. N., Lim, J., Villiger, L., ... & Gootenberg, J. S. (2023). Drag-and-drop genome insertion of large sequences without double-strand DNA cleavage using CRISPR-directed integrases. Nature Biotechnology, 41(4), 500-512.

Getting started

pegRNA_rank has been modified to accept the use of amplicons and inserts directly from the command-line. This provides a convenient way to run and increase the throughput of the original pegRNA_rank.py code.

Providing a single amplicon and insertion sequence

pegRNA_rank now accepts user inputs for amplicons (--amplicon) and inserts (--insert). The current version of pegRNA_rank can only accept a single amplicon and insert per execution.

python3 rank.py --amplicon <amplicond sequence> --insert <insert sequence>

Providing a list of amplicons and insertion sequences

pegRNA_rank has been modified with the additon of auto_pegRNA_rank.sh. This bash script allows users to provide a separate file containing all of the amplicons that you want to test and the insertion sequences that you want to screen. Each sequence needs to be on a new line, and the amplicons and insertions must be in different files.

auto_pegRNA_rank.sh currently requires both a amplicon and insertion file to be presented to function.

auto_pegRNA_rank.sh --amplicon <amplicon file> --insert <insertion file>

Installing

Prerequisites

The current pipeline has been tested on the following package versions

  • Python 3.6 or later is required
  • numpy 1.26.2
  • pandas 2.1.3
  • torch 2.2.1

pip installation

pegRNA_rank instructions: Make sure you have Python>=3.6 and PyTorch>=1.5 installed. Then, install dependencies with:

pip install -r requirements.txt

If you need to type python3 to use python and you are having problems with dependencies not being available, please consider using pip3

pip3 install -r requirements.txt

Alternative method of running pegRNA_rank

When you only have a single insert or amplicon that you want to screen across a much larger library, consider using the the following command to read and loop each line within a text file.

Example 1

while IFS= read -r amplicon; do python3 rank.py --amplicon "$amplicon" --insert ATGATGCCATGCTGATCGATGCTGAC; done < sequences.txt

In the following example, the command will loop across each line in sequences.txt ($amplicon) and use the insertion sequence ATGATGCCATGCTGATCGATGCTGAC.


Example 2

while IFS= read -r insert; do python3 rank.py --amplicon ATGATGCCATGCTGATCGATGCTGAC --insert "$insert"; done < insertion.txt

In the following example, the command will use the amplicon sequence ATGATGCCATGCTGATCGATGCTGAC and loop line by line the file insertion.txt for the insertion sequences to use.

About

A modified version of Deep learning MLP model to rank pegRNA sequence

Resources

Stars

Watchers

Forks

Releases

No releases published

Packages

No packages published

Languages

  • Python 83.9%
  • Shell 16.1%