bigr_rna_editing

This pipeline uses SPRINT and RNAEditingIndexer to identify editing events from paired-end RNA-seq data.

Installation

Gustave Roussy users

The pipeline is already installed on the Flamingo cluster of Gustave Roussy.
It is localized here: /mnt/beegfs02/pipelines/bigr_rna_editing/

Admin : installation of a new version of the pipeline

1️⃣ Download pipeline

cd /mnt/beegfs02/pipelines/bigr_rna_editing/
VERSION="1.1.1"
git clone https://github.com/gustaveroussy/bigr_rna_editing.git ${VERSION}

2️⃣ Download Singularity images

Download all singularity images from Zenodo:

cd /mnt/beegfs02/pipelines/bigr_rna_editing/${VERSION}/envs/singularity/
wget https://zenodo.org/api/records/14916660/files-archive
unzip files-archive

3️⃣ Install Snakemake environment

source /mnt/beegfs02/software/recherche/miniconda/25.1.1/etc/profile.d/conda.sh
conda env create -f /mnt/beegfs02/pipelines/bigr_rna_editing/${VERSION}/envs/conda/snakemake.yaml --prefix=/mnt/beegfs02/pipelines/bigr_rna_editing/${VERSION}/envs/compiled_conda/snakemake -y

You are now ready to use the pipeline!

Using

You need to make 2 files: a design file and a configuration file.

Configuration file

• design: absolute path to your design.csv file.
• output_dir: absolute path to the output directory where results will be saved.
• reference: the reference to use for the alignment and the idetification of editing events. Possible choices are "hg19", "hg38", "mm10" or "mm9". The reference will be downloaded from the UCSC web site.
• samples_order_for_ggplot (optional): the order of samples for the x axis of graphs (you can order samples by condition for example). Default is alphabetical order.
• SPRINT_extra (optional): extra parameters for "SPRINT main" command.
• RNAEditingIndexer_extra (optional): extra parameters for "RNAEditingIndexer" command.
• nb_sampled_reads (optional): number of reads to sample. Possible choices are "" for no sampling, "50000000" for 50M of reads (can be another integer), or "auto" (to sample the minimum number of reads obtain throught all samples if they have more than min_nb_sampled_reads_for_auto option, else the threshold is the value of min_nb_sampled_reads_for_auto. If a sample has less than the threshold, all its reads are used.
• min_nb_sampled_reads_for_auto (optional): minimum number of reads to sample if nb_sampled_reads is set to "auto". Defaut is 50M of reads.

Example:

design: "/mnt/beegfs02/scratch/m_aglave/Editing_analysis/script/design.csv"
output_dir: "/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_output/"
reference: "hg38"
samples_order_for_ggplot: "S1_patient,S3_patient,S2_patient"
SPRINT_extra: ""
RNAEditingIndexer_extra: ""

Design file

It must be a comma separated file (.csv where comma is ",") with 3 columns:

• sample_id: the sample name of you sample (it could be different that your fastq files).
• R1_fastq: absolute path to the R1.fastq.gz file.
• R2_fastq: absolute path to the R2.fastq.gz file.

Example:

sample_id,R1_fastq,R2_fastq
S1_patient,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S1-patient_R1.fastq.gz,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S1-patient_R2.fastq.gz
S2_patient,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S2-patient_R1.fastq.gz,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S2-patient_R2.fastq.gz
S3_patient,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S3-patient_R1.fastq.gz,/mnt/beegfs02/scratch/m_aglave/Editing_analysis/data_input/S3-patient_R2.fastq.gz

Notes:

• sample names mustn't contain special characters or spaces.
• fastq files must be gzipped.

Run

You need snakemake (via conda) and singularity (via module load). They are already installed for you on Flamingo, just follow the example below.
Don't forget to change the version of the pipeline and the path to your configuration file.
Example of script:

#!/bin/bash
#using: sbatch run.sh
#SBATCH --job-name=Editing_analysis
#SBATCH --nodes=1
#SBATCH --cpus-per-task=1
#SBATCH --mem=250M
#SBATCH --partition=longq

source /mnt/beegfs02/software/recherche/miniconda/25.1.1/etc/profile.d/conda.sh
conda activate /mnt/beegfs02/pipelines/bigr_rna_editing/<version>/envs/compiled_conda/snakemake
module load singularity-ce
Editing_pipeline="/mnt/beegfs02/pipelines/bigr_rna_editing/<version>/"

snakemake --profile ${Editing_pipeline}/profiles/slurm \
          -s ${Editing_pipeline}/Snakefile \
          --configfile <path_to/my_configuration_file.yaml>

Steps of the pipeline

1. Symbolic link of fastq files
2. Reads QC & Trimming (fastQC, fastp & multiqc)
3. BWA index generation (via SPRINT)
4. Sampling read (optional) (seqtk)
5. BWA alignement (via SPRINT)
6. Aligment QC (Samtools & multiqc)
7. Identification of Editing events (SPRINT) (this step takes 2-3 days!)
8. Summary of SPRINT results (R)
9. Bam sorting (Samtools)
10. Identification of Editing events (RNAEditingIndexer)
11. Summary of RNAEditingIndexer results (R)

Information about Editing tools:
SPRINT:
https://github.com/jumphone/SPRINT
https://academic.oup.com/bioinformatics/article/33/22/3538/4004872
RNAEditingIndexer:
https://github.com/a2iEditing/RNAEditingIndexer
https://pubmed.ncbi.nlm.nih.gov/31636457/