This project tries to re-animate the project from Oxford Nanopore. CURRENTLY IT IS NOT IN A WORKING STATE. Please see their current nextflow implementation as a reference wf-transcriptomes, which contains functionality for differential expression.
Pipeline for differential gene expression (DGE) and differential transcript usage (DTU) analysis using long reads
This pipeline uses snakemake, minimap2, salmon, edgeR, DEXSeq and stageR to automate simple differential gene expression and differential transcript usage workflows on long read data.
If you have paired samples (e.g for example treated and untreated samples from the same individuals) use the paired_dge_dtu branch.
The input files and parameters are specified in config.yml
:
transcriptome
- the input transcriptome.annotation
- the input annotation in GFF format.condition_a_identifier
- a string identifiying the first trait.condition_b_samples
- a string identifiying the second trait.alignments/*.bam
- unsorted transcriptome alignments (input to salmon
).alignments_sorted/*.bam
- sorted and indexed transcriptome alignments.counts
- counts generated by salmon
.merged/all_counts.tsv
- the transcript count table including all samples.merged/all_counts_filtered.tsv
- the transcript count table including all samples after filtering.merged//all_gene_counts.tsv
- the gene count table including all samples.de_analysis/coldata.tsv
- the condition table used to build model matrix.de_analysis/de_params.tsv
- analysis parameters generated from config.yml
.de_analysis/results_dge.tsv
and de_analysis/results_dge.pdf
- results of edgeR
differential gene expression analysis.de_analysis/results_dtu_gene.tsv
, de_analysis/results_dtu_transcript.tsv
and de_analysis/results_dtu.pdf
- results of differential transcript usage by DEXSeq
.de_analysis/results_dtu_stageR.tsv
- results of the stageR
analysis of the DEXSeq
output.de_analysis/dtu_plots.pdf
- DTU results plot based on the stageR
results and filtered counts.conda
.conda
.conda
feature of snakemake
.README.md
Snakefile
- master snakefileconfig.yml
- YAML configuration filesnakelib/
- snakefiles collection included by the master snakefilelib/
- python files included by analysis scripts and snakefilesscripts/
- analysis scriptsdata/
- input data needed by pipeline - use with caution to avoid bloated reporesults/
- pipeline results to be commited - use with caution to avoid bloated repoClone the repository:
git clone https://github.com/snakemake-workflows/transriptome-differential-expression
Edit config.yml
to set the input datasets and parameters then issue:
On a server, e.g.:
snakemake --use-conda -j <num_cores> all
On a cluster, e.g.
snakemake --slurm --default-resources slurm_account=<your slurm account> slurm_partition=<your clusters default partition> -j <unlimited or lower> --configfile ./envs/<your config yaml> --workflow-profile ./profile/ --snakefile <path to Snakefile> --directory <desired working directory>
Note, that the profile offers a template cluster configuration - it needs adjusting for particular clusters. Contributions of particular configurations are welcome!
(c) 2018 Oxford Nanopore Technologies Ltd. (c) 2023- Lukas Hellmann & Christian Meesters (JGU Mainz, Germany)
This Source Code Form is subject to the terms of the Mozilla Public License, v. 2.0. If a copy of the MPL was not distributed with this file, You can obtain one at http://mozilla.org/MPL/2.0/.
This worflow is largely based on the approach described in the following paper: