Snakemake pipeline for alternative splicing analysis with VAST-TOOLS.

Table of Contents:

• Getting started
  • Requirements
  • Installing
  • Running sm-rnaseq-splicing
• Output
• Authors
• License

sm-rnaseq-splicing requires the following:

• VAST-TOOLS and its dependencies (environment module)
• Snakemake >= 7.22 (environment module)
• Mamba >= 23.1.0 (environment module)
• Singularity (environment module)
• Python >= 3.6

Clone this repo git clone https://github.com/simon-bt/sm-rnaseq-splicing.git. The repo contains the structure required for the pipeline execution.

1. Modify config/samples.tsv file to provide sample and group information.
2. Modify config/config.yaml file to specify the type of reads and parameters used for alternative splicing analysis and downstream data visualisation.
3. Modify profile/config.yaml file to better suit your SLURM scheduler configuration and assign per-rule job requirements.
4. Provide soft-link paths to FASTQ files in {1|2}_fastq.gz format in data/ folder.
5. Modify wrapper.sh file to better suit your SLURM scheduler configuration.
6. Modify path of 'singularity-args' in profile/config.yaml to bind root directory using Singularity.

It is recommended to execute a dry run from the top directory before submitting wrapper.sh to the scheduler: snakemake --snakefile Snakefile --slurm --profile profile/ --dry-run

sm-rnaseq-splicing pipeline uses vast-tools to quantify alternative splicing changes from RNA-sequencing data. The pipeline aligns sequencing reads to a database for a given species, combines vast-tools output into inclusion table and allows to compare two conditions at a time of runtime. The comparisons can be performed for any given two conditions upon their specification in the config/config.yaml file. Pipeline can be also re-run with a different dPSI threshold.

vast_compare and visualize_data rules generate:

1. In results/{GroupB}_vs_{GroupA}/data:

• DataAll-*.tsv files - inclusion levels for splicing events (all and separately) passing coverage and other default criteria. Events passing criteria for the regulation are indicated.
• DataDiff-*.tsv files - more extended inclusion level data for the regulated splicing events (all and separately)

visualize_data rule generates figures for microexons (exons shorter than 28 nt), exons (longer than 27 nt) and introns:

2. In results/{GroupB}_vs_{GroupA}/figures/splicing/ (pdf format):

• Scatter_* - scatter plot of splicing quantification levels between two conditions for a given event type
• DistDPSI - box distribution of dPSI values for the regulated events
• LevelsDiff - violin distribution of splicing quantification levels for the regulated events
• RidgeDPSI - ridge density plot for changes in splicing quantification levels for the regulated events
• ProteinImpact - stacked % bar plot of the protein impact predictions for the regulated events
• HMapLevels_Standardized - heatmap of splicing quantification levels for the regulated events, standardized to 1
• HMapLevels_ZScores - heatmap of splicing quantification levels for the regulated events, in z-score of quantification values
• PercentageDPSI - stacked % bar plot of the proportion of truly regulated events that pass vast-tools compare criteria to all events that pass the coverage filter

3. In results/{GroupB}_vs_{GroupA}/figures/regulated_events/ (pdf format) - per-sample quantification level plot for the regulated microexons (exons shorter than 28 nt), exons (longer than 27 nt) and introns

prepare_sashimis and plot_sashimis rules generate sashimi plots for the regulated microexons (exons shorter than 28 nt), exons (longer than 27 nt) and introns:

4. In results/{GroupB}_vs_{GroupA}/figures/sashimis/ (pdf format) - configuration files and sashimi plots

Simon Bajew, PhD (IIS Biodonostia)

This project is under MIT License.