RNA_seq analysis pipeline

The scripts and pipelines used to analyze data for () manuscript.

Docker images that are called in this script can be found here:

• general_docker (Version 1)
• r_docker (Version 1)

Running the pipeline

A helper script to create the sample sheet and to start creating the rmats and deseq sample info can be run with

python src/scripts/make_sample_sheets.py

A few notes about this script

1. The default fastq directory is raw_data, however you can provide your own with -d.
2. The default output directory is files, however you can provide your own with -o.
3. The samples.tsv file will be fully ready to use with the snakemake pipeline.
4. The sample_info.csv file will be made with samples and columns added in, but you will need to fill it out with your own sample information.
5. The rmats_sample_info.tsv file will be made with samples and columns added in, but you will need to fill it out with your own sample information.
6. A warning is printed when running the script to remind you that these two files are not complete.

Snakemake

A snakemake pipeline that can be used to run bulk RNA-seq analysis. Can chose between cutadapt, bbduk or no adapter trimming. Outputs fastqc summary files, star summary files, and a counts matrix, and an html file containing quality control images. All of the output files can be analyzed using the rmd script

Writen by Kristen Wells

To use:

1. Download and install miniconda3: For Linux

wget https://repo.continuum.io/miniconda/Miniconda3-latest-Linux-x86_64.sh bash Miniconda3-latest-Linux-x86_64.sh

2. Install Snakemake:

conda install snakemake -c bioconda -c conda-forge

This pipeline has been configured to run on a slurm scheduler. For more information on running snakemake on different cluster configerations see the executor help page

Changing the profile should be the only major update to move to a different scheduler, however, I have added information for custom log files to each rule that is specific to slurm.

3. Update the config file (config.yaml)

• SAMPLE_TABLE: path to a file consisting of at least two columns.
  • The first column should be titled sample and contain the name of the sample.
  • The second should be titled fastq1 and contain the path to the associated fastq file
  • The third optional column should be titled fastq2 and be used if you have paired end data. This should contain the path to the associated read 2 fastq file.
  • The fourth optional column is named spike-in with TRUE/FALSE values per sample indicating if spikeins were used. If this column is not included it will be assumed that spike-ins were not included.
• PROJECT: The name of the project, will be the name of the output counts matrix
• GENOME: The path to the genome directory created by star
• COMBINED_GENOME: Optional, only include this if spike-ins were used. This is the genome of both the main organism and spike-in organism
• GTF: The path to the GTF associated with the star genome
• COMBINED_GTF: Optional, only include with spike-ins, the path to the combined GTF file
• RESULTS: The path to the results directory
• ADAPTORS: Optional, only include if using bbduk for adaptor trimming. Path to the adaptors file in the bbtools package
• TRIM_METHOD: What trim method to use. Can be "bbduk", "cutadapt", or "no"
• PE and SE: extra paramamters for all of the jobs run

4. Update snakecharmer.sh to your specific cluster specs.

5. submit the job using sbatch snakecharmer.sh

R analysis

The file src/scripts/DESeq_analysis.Rmd was used for RNA-seq analysis. Figures were made with src/scripts/figures.R