seq_pipeline

Running the pipeline

The pipeline is run using conda environments on the local machine using:

snakemake --cores {cores} --resources ncbi_connection=1 --use-conda --conda-frontend mamba all

Config File config.yml and Metadata File metadata.tsv

For examples of different config files see parameter_templates/

Aligners

The pipeline supports either STAR or bowtie2 for aligning given in config.yml as aligner: star or aligner: bowtie2

Input Files

Currently the pipeline only acceps pair end data, except for bam file inputs.

It expects readlength: {the_readlength} in config.yml and optionally for aligning adapters can be specified in config.yml see parameter_templates/adapter_trimming.yml, with type of adapter present beind passed on to cutadapt.

SRR

With metadata_files: srr in config.yml and SRRxxxxxxxx for sample_ids in metadata.tsv

fastq

With metadata_files: fastq in config.yml and R1 and R2 columns with paths to the respective fastq files in metadata.tsv

bam

With metadata_files: bam in config.yml and bam column with paths to the respective bam files in metadata.tsv

Pipelines

Bulk

Runs qc, and generates a count matrix, super basic ready for downstream analysis.

Ripseq

The pipeline runs peakcalling using MACS2, and then run through various peak detection methods:

• IDR
• PePr
• DEQ
• Thor
• Genrich

before saving pileup summaries of these analysis tools. The pipeline strictly requires unique matched input control for each condtion.

The metadata.tsv file requires one line per sample with the following columns:

• sample_id
• condition
• method: IP or Input
• matching_input_control: the sample_id of the matching input control

The config file requires the following terms in config.yml:

• control_condition: setting which condition is compared against, all other conditions compared against this

See parameter_templates/ripseq_config.yml and parameter_templates/ripseq_metadata.tsv for an example template

Stamp

The pipeline runs the Bullseye C to T editing pipeline an performs analysis on the data, and plotting pileups with optionally other bam files on edit genes.

The metadata.tsv file requires one line per sample with the following columns:

• sample_id
• condition
• method: IP or Input
• matching_{condition}: One column per condition providing the matching samples of the same condition

The config file requires the following terms in config.yml:

• complex_comparisons: allowing comparisons of one condition against multiple conditions
• simple_comparisons: allowing comparisons of pairs of conditions
• display_order: determining the order of conditions in the plots

See parameter_templates/stamp_config.yml and parameter_templates/stamp_metadata.tsv for an example template

Dependencies

These will be downloaded automatically by the pipeline when running using conda.

Base

• snakemake
• sra-tools
• parallel-fastq-dump
• samtools
• [bedtools](http://bioinformatics.oxfordjournals.org/content/26/6/841.short
• pybedtools
• tidyverse
• pandas
• STAR
• bowtie2
• cutadapt
• fastqc
• multiqc
• biomaRt
• GenomicRanges
• Rsamtools
• rtracklayer
• ggseqlogo

STAMP

• BULLSEYE

Ripseq

• MACS2
• IDR
• piranha
• genrich
• THOR
• PePr
• DEQ modified to run on more modern versions of R