Introduction of MOFF

                            # # # # #   # # # #   # # # #  # # # # 
                            #   #   #   #     #   #        #       
                            #   #   #   #     #   # # # #  # # # #    
                            #   #   #   #     #   #        #          
                            #   #   #   # # # #   #        #          
                                 

Hi, my name is MOFF. I was designed to predict off-target effects for CRISPR/Cas9! 

I have three major functions: 1). Predict off-target effects for any given gRNA-target pair.
                              2). Predict genome-wide off-target effects for any sgRNA.
                              3). Select best sgRNAs for allele-specific knockout.


Hope you enjoy playing with me ^o^!
                                 
Any questions or bugs, please contact hwkobe.1027@gmail.com or whe3@mdanderson.org

How to install MOFF

Requirements

• Python>=3.4
• pandas, numpy, scipy
• tensorflow, keras

Installation

Step1: Install Anaconda (highly recomended)

wget https://repo.continuum.io/archive/Anaconda2-2018.12-Linux-x86_64.sh 
bash Anaconda2-2018.12-Linux-x86_64.sh 

Step2: Install MOFF through pip

pip install MOFF

Step3: OR you can install MOFF through git clone

git clone https://github.com/MDhewei/MOFF.git
cd MOFF
python setup.py install

How to use MOFF

1. MOFF score: Predict off-target effects for given gRNA-target pairs

Inputs for MOFF score

MOFFscore requires the user to provide .csv or .txt file containing sgRNA sequences and corresponding DNA target sequences. Each line should have one gRNA(20bp+PAM) and one target(20bp+PAM) sequence. Note that MOFF is designed for mismatch-only off-target prediction, not for indel mutations. See example file here.

 Arguments of the program:

 -i/--inputfile (required): 
 Input files containing sgRNA sequences and corresponding DNA target sequences.

 -p/--prefix (Optional): 
 Prefix of the file to save the outputs, default: ScoreTest.

 -o/--outputdir (Optional): 
 Directory to save output files, if no directory is given a folder named MOFF_scores
 will be generated in current working directory.

Example to run MOFFscore

MOFF score -i MOFFscore_test.txt

Columns of Output table

- crRNA: the sgRNAs designed to target specific DNA sequence
- DNA: the DNA sequence of the off-target site 
- MDE: predicted mismatch-dependent off-target effect 
- CE:  predicted combinatorial effect
- MMs: the number of mismatches between sgRNA and off-target
- GMT: predicted guide-intrinsic mismatch tolerence 
- MOFF: the final MOFF score predicted for given gRNA-target pair

2. MOFF aggregate: Predict the genome-wide off-target effects for given sgRNAs

MOFF aggregation can directly take the outputs of CRISPRitz and Cas-OFFinder as inputs. Output table files generated by any genome-wide off-target searching methods are also supported in theory, but the columns of outputs for different methods are different, therefore, it is required to modify the column name of sgRNA(20bp+PAM) and target(20bp+PAM) to 'crRNA' and 'DNA' respectively. Note that MOFF only supports mismatch-only off-target predictions, indel mutations are not applicable. File formats including .csv and .txt are accepted. See examplefile here.

 Arguments of the program:

 -i/--inputfile (required): 
 Input files containing all the potential off-target sites in the genome for given sgRNA(s).

 -p/--prefix (Optional): 
 Prefix of the file to save the outputs, default: AggregationTest.

 -o/--outputdir (Optional): 
 Directory to save output files, if no directory is given a folder named MOFF_aggregation
 will be generated in current working directory.

Example to run MOFF aggregate

MOFF aggregate -i MOFFaggregation_test.txt

Columns of Output table

- sgRNA: the sgRNAs selected to predict genome-wide off-target
- GMT: aggregated guide-intrinsic mismatch tolerence 
- MDE.sum: aggregated mismatch-dependent off-target effect 
- MOFF.sum: aggregated MOFF score for specific sgRNA