finding ORIs in stranded SNS-seq

This repository contains scripts and files used to analyze stranded paired-end SNS-Seq experiments for Stanojcic et al in prep.

The scripts are made to run on a HPC under slurm (Roscoff Bioinformatics platform ABiMS (http://abims.sb-roscoff.fr)).

The first script qc_mapping.sh is for read quality control and mapping if you have already mapped reads you can directly start with the second script finding_ORIs.sh, that will used aligned reads from stranded SNS-seq as input.

The first script qc_mqpping.sh uses raw paired-end stranded sequencing reads from SNS-seq experiments and performs quality check, adapter and quality trimming and read mapping.

• quality control of the provided fastq files (fastqc 0.11.9)
• trimming of tail and adapter sequences (cutadapt 4.0)
• quality trimming with q20 threshold (trimmomatic 0.39)
• alignment against provided indexed genome (bowtie2 2.4.1)
• sorting and conversion fom sam to bam of aligned reads (samtools 1.13)
• check for insert size (picard 2.23.5)
• removal of duplicated reads (picard 2.23.5)
• bam file indexing (samtools 1.13)
• generation of Multiqc report (multiqc 1.13)

The second script finding_ORIs.sh will use paired end aligend reads (sorted bam) as input and outputs bed files of the detected ORIs

• seperation of mapped reads in minus and plus strand (samtools 1.13)
• strand-seperated peak calling (macs2 2.2.7.1)
• peak filtering (bedtools 2.2.0.0 and awk) based on three criteria:
  • distance of minus and plus strand peaks
  • no complete overlap between plus and minus strand peaks
  • right strand orientation (minus followed by plus)

Parameters

Parameters for the qc_mqpping.sh script need to be provided in the config_mapping.txt file, that has to be in the same working directory as the script :

• output directory name. The directory will be generated by the sript in the working directory.

  output_dir=alignedReads

• list of sample names as found in the fastq files e.g. $sample\L001_R1_001.fastq.gz

  input_list=("x_S1_" "y_S2_" "z_S5_")

• path to fastq directory

  fastq_directory=path/to/fastqfiles

• path and genome prefix of bowtie2 index and genome.fasta

  genome=path/to/genome/myfavorite_Genome

• genome prefix for filenames of mapped reads

  genome_prefix=myfavorite_Genome

Parameters for the finding_ORIs.sh script need to be provided in the config_finding-ORIs.txt file, that has to be in the same working directory as the script:

• path to aligned reads

  read_directory=path/to/aligned_reads

• output directory name (generated by the sript in the working directory)

  output_dir=results/myresults

• prefix of mapped reads ($sample$prefix.bam)

  file_prefix=aln-pe_genomeprefix_sorted_reheadered_dups-removed

• sample name as found in bam files

  input_list=("x_S1_" "y_S2_" "z_S5_")

• peak overlap to be excluded in percent

  overlap_list=(50)

• window size for peak pair selection in nucleotides

  window_list=(500)

Run

The scripts are written to be run on an HPC cluster under slurm

  sbatch findingORIs/qc_mapping.sh

  sbatch findingORIs/finding_ORIS.sh

Authors

• @bridlinbarckmann