It is specifically designed for Bazzini Lab at Stowers Institute for Medical Research.
There are 3 submodules in this pipeline
- design
- mismatch
- all
The design module designs gRNAs from an input fasta file. This modules runs RNAplfold and computes a score using method designed by Ariel Bazzini
- Run RNA fold for each gene (Complete sequence). RNA fold takes a fasta sequence as input along with the length of gRNAs.
- Take sliding window of gRNAs for a specific length from gene fasta. (The length of gRNA will be equal length from step 1. e.g., 20, 22, 23)
- Compute the mean score of all columns (The number of columns will be equal to length of the gRNA) in each row. This gives one value for each bp of a gene.
- For each sequence from step 2, take the scores from step 3 (The number of scores should be equal to the length of the gRNA sequence). Compute a mean of these scores to give one single score for each gRNA
Input file
Fasta File: The input file is a FASTA file containing sequences for which gRNAs should be designed.
The mismatch module is used to compute the number of mismatches for gRNAs against the transcriptome. The module runs water pairwise aligner from Emboss.
Input file The input files are:
- gRNA File: The gRNA file can be provided in one of the following formats:
- FASTA format: The file should end with
.faor.fasta, with the FASTA headers representing the gRNA names. - Tab-delimited format: The file should end with
.txt, with the first column containing the gRNA names and the second column containing the gRNA sequences. The file must include headers.
- FASTA format: The file should end with
- Transcript/cDNA FASTA File: A FASTA file containing transcript or cDNA sequences.
The 'all' module runs both the 'design' and 'mismatch' module
Input file The input files are:
- Fasta File: The input file is a FASTA file containing sequences for which gRNAs should be designed.
- Transcript/cDNA FASTA File: A FASTA file containing transcript or cDNA sequences.
This pipeline runs in a conda environment and all the required packages are installed in the conda environment.
However additional tools mentioned below are required for running the pipeline. Make sure you have them installed and in path.
Additional tools
The mismatch module uses water aligner from EMBOSS. The installation instructions are mentioned in the links above.
To run it on Stowers HPC, Emboss module is already installed and can be loaded using
ml emboss/6.6.0Create a Conda environment using the yaml file
conda env create -f gRNADesign.yml
Activate the gRNADesign environment
conda activate gRNADesign
Check the different modules in the pipeline
python gRNADesign.py --help
usage:
Script to design gRNA's using RNAplfold and formula designed by Ariel Bazzini
Optional Arguments:
-h, --help show this help message and exit
Subcommands:
{all,design,mismatch}
all Arguments to run full pipeline
design 'design' specific arguments
mismatch 'mismatch' specific arguments
For any questions Contact Huzaifa Hassan: [email protected]
To check the arguments for each module, run
python gRNADesign.py <submodule name> --help
python gRNADesign.py design --help
python gRNADesign.py mismatch --help
python gRNADesign.py all --help
usage: all [-h] -f FASTA -t TRANSCRIPT_FASTA -o OUTPUT_FILE [-n TOP_N]
[-l LENGTH] [-m MISMATCHES] [-p NUM_PROCESSES]
Arguments:
-h, --help show this help message and exit
-f FASTA, --fasta FASTA
Fasta file with sequences to design gRNA
-t TRANSCRIPT_FASTA, --transcript_fasta TRANSCRIPT_FASTA
cDNA fasta file
-o OUTPUT_FILE, --output_file OUTPUT_FILE
Output CSV file
-n TOP_N, --top_N TOP_N
The number of sequences to write to file (based on
highest score) (Optional). Default is top 10
-l LENGTH, --length LENGTH
Length of gRNAs to design (Optional). Default is 23
-m MISMATCHES, --mismatches MISMATCHES
The number of mismatches to report in a separate file.
Default is 5
-p NUM_PROCESSES, --num_processes NUM_PROCESSES
The number of processes/threads to use. Default is 1
These are examples for each module
python gRNADesign.py all -f gene.fasta -t cDNA_transcript.fasta -o gRNA.csv -n 10 -l 22 -m 5
python gRNADesign.py design -f gene.fasta -o gRNA.csv -n 10 -l 22
python gRNADesign.py mismatch -g gRNA.csv -t cDNA_transcript.fasta -m 5 -p 4
The results are written to a folder 'results'
results/grna : The gRNAs designed are in this folder
results/all_alignments : The alignments of gRNA to the transcripts are in this folder
results/mismatches : The alignments of gRNA to the transcripts having mismatches less than or equal to the specified mismatches.
The results/grna will have one file with the name used while running the pipeline (eg gRNA.csv)
The columns of gRNA.csv are described below:
- gRNA_name : The name here will be the sequence id of the input sequences followed by the start and the end coordinate where this gRNA is from on the sequence.
- seq_23nt : The length of the gRNA. Default is 23, unless user defined.
- Gene|Transcript : The name here will be the sequence id of the input sequences
- Seq_start : The start coordinate where this gRNA is from on the sequence
- Seq_end : The end coordinate where this gRNA is from on the sequence
- Average_score : Average is the score defined in the design module
- G_Percentage : Percentage of Base 'G' in the gRNA
- C_Percentage : Percentage of Base 'C' in the gRNA
- A_Percentage : Percentage of Base 'A' in the gRNA
- T_Percentage : Percentage of Base 'T' in the gRNA
The results/all_alignments and results/mismatches will have one file for each gRNA. results/all_alignments will have all the alignments of gRNA mapping to each gene/transcript from the TRANSCRIPT_FASTA file and results/mismatches will have the same alignments filtered for gene/transcript with total mismatch of specified mismatches.
- The columns of all_alignments.csv (eg all_alignments_gRNA_1.csv) are described below:
- Sequence ID : The ID of sequence from the input TRANSCRIPT_FASTA file
- gRNA_align_start : Alignment start of the gRNA on the gene/transcript
- gRNA_align_end : Alignment end of the gRNA on the gene/transcript
- seq_align_start : Alignment start of the gene/transcript
- seq_align_end : Alignment end of the gene/transcript
- identity : Number of aligned bases including gaps in gRNA
- total_aligned_len : Number of bases which aligned between the two sequences
- missing_start : The number of unaligned bases at the start of gRNA
- missing_end : The number of unaligned bases at the end of gRNA
- added_aligned_len : This is the total_aligned_len + missing_start + missing_end (Which represents a hypothetical aligned + unaligned length)
- gRNA_gaps : The number of gaps in the gRNA for this alignment
- seq_gaps : The number of gaps in the gene/transcript for this alignment
- total_mismatches : This is added_aligned_len - identity ( Which is the difference of aligned bases and the total alignment including the unaligned length)
Note: Please feel free to compute your our own mismatch score from files in results/all_alignments if needed.
This page is still being update
Helper Scripts
- Filter gRNAs based on the distance between the gRNAs
This script is used to select gRNAs based on the distance between the gRNAs. The script takes the gRNA.csv file as input and the distance between the gRNAs. The script will select the gRNAs based on the distance and write the selected gRNAs to a new file.
Usage:
python select_gRNA_dist.py -h
usage: select_gRNA_dist.py -f <df_fh> -d <dist_grna> -n <top_n>
Selects the top n rows based on 'Average_score' within each group of 'Gene|Longest_transcript' column, while ensuring that the
gRNAs are at least dist_grna apart.
options:
-h, --help show this help message and exit
-f DF_FH, --df_fh DF_FH
File path of the input CSV file.
-d DIST_GRNA, --dist_grna DIST_GRNA
Minimum distance between 'Seq_start' values of selected rows. Default is 30.
-n TOP_N, --top_n TOP_N
Number of top rows to select within each group. Default is 10.