psiFinder_snakemake

snakemake pipline for Ψ-sites Identification/Annotation/Target

psiFinder_snakemake is a snakemake pipline (back-end code) correspond to psiFinder quick start QT widget (front-end user interface).

Contents

• Pre-installation
• Input data and required files
• Usage
  • Set configuration
  • Run psiFinder snakemake

Pre-installation

psiFinder_snakemake requires snakemake/seqtk/cutadapt/STAR/bedtools/gtfToGenePred and several perl/python/R packages pre-installation and predominantly used in unix-based operating systems. Therefore, for the usability of psiFinder_snakemake, we recommend running all the tools and scripts in WSL2 (WSL2 installation guide: https://pureinfotech.com/install-windows-subsystem-linux-2-windows-10/) or unix-based system with perl, python and R.

Required perl modules:

cpan Getopt::Long;
cpan Bio::Perl;
cpan Bio::SeqIO;

Required python modules:

pip install pandas
pip install matplotlib
pip install numpy
pip install pickle
pip install scikit-learn
pip install argparse

Required R packages:

#use pacman to install packages in batch
install.packages("pacman")
library(pacman)

#load and install required R packages 
p_load("optparse","devtools","caTools","neuralnet","NeuralNetTools","dplyr","stringr","gridExtra","cowplot","pROC","mccr","ggplot2","ggpol","ggpubr","RColorBrewer","openxlsx","reshape2","factoextra","bedr","scales","e1071","tidyr")

Input data and required files

Test data and big files for the snakemake:

• Input (CMC-control) file for Ψ-sites identification: A1.cutadapt.extendedFrags.collapse.cutBarcodes.fa.gz (download from: https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw)
• Treat (CMC-treated) file for Ψ-sites identification: A2.cutadapt.extendedFrags.collapse.cutBarcodes.fa.gz (download from: https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw)
• Genome files for STAR alignment: hg38.fa, hg38.fa.fai (download from https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw and deposit it in ./snakemake/genome)
• Annotation file for Ψ-sites identification: hg38.gencode.v30.tRNA.refseqNcRNA.geneAnno.bed12 (download from https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw and deposit it in ./snakemake/script)
• Annotation file for Ψ-sites annotation: hg38.genecode.v30.tRNA.snoRNA.miRNA.rmsk.exonFeatures.bed6 (download from https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw and deposit it in ./snakemake/script/annotation)
• Annotation file for Ψ-sites metagene: gencode.v32.chr_patch_hapl_scaff.annotation.gtf (download from https://mega.nz/folder/oaUmhK7I#gSuYH4HW7OhL5qEbgmw0fw and deposit it in ./snakemake/script/metagene)

Usage

Set configuration

psiFinder_snakemake require a quickstart_config.yml file to get user option and start the running:

Run psiFinder snakemake

bash run_psiFinder_snakemake.sh

Or use psiFinder quick start QT widget to interactively run psiFinder_snakemake:

Notice: If Ψ-sites Target of psiFinder quick start QT widget is selected, psiFinder_snakemake will use human_hg38_snoRNABase_snoDB_rmRepeat.collapse.fa (ACAscan required in ./snakemake/script/target_prediction, predict H/ACA snoRNA-Ψ-target interaction, known H/ACA snoRNA sequence) and overall_multinomialnb_model.pkl (PUSscan required in ./snakemake/script/target_prediction, predict PUS-Ψ-target interaction, pre-built model specialized for TRUB1/PUS3/PUS1 target prediction [can be generated by psiMNB workflow: https://github.com/chenzhr23/psiMNB]).