RNA-Seq workflow

RNA-Seq workflow#

Let’s run a commonly used Nextflow pipeline to analise RNA-seq data (nf-core/rnaseq).

This pipeline that we will run will cover the following 4 steps using diffrent coomonly used bioinformatics tools:

  1. Indexes a transcriptome file

  2. Quality control of the high throughput sequence data with FastQC

  3. Quantification of transcripts using Salmon

  4. Creates a MultiQC report. MultiQC searches a given directory for analysis logs and compiles a easy-viwing HTML report.

This is based on this part of Nextflow tutorial. There you can find the pipeline broken into the different pieces in a collection of 7 different scripts if you want to get into the details.

To be able to run this pipeline we will clone the training repository from Nextflow training materials in github. Here you can enjoy the hello training, their fundamentals training based on the rnaseq pipeline and the advanced training with all data, files and scripts needed.

First cloning with git clone:

git clone https://github.com/nextflow-io/training.git --depth 1

And then move to training/nf-training:

cd training/nf-training

Here there are a colloection of 7 scripts where the rnaseq pipeline (a small version of it) is broken into the different parts. It is great training material and I recommend you go through if you want to keep learning Nextflow. For today we will run the last script that contains all the parts we want to run (Indexing transcriptome file, fastQC, Salmon for quantification and MultiQC to see a final report of our run).

This pipeline requires Docker as every process is run inside a Docker container. On other words an environment where the required operation system and software is installed with all dependencies. Nextflow uses generally containerization tools like Docker, Singularity, or Conda to ensure reproducibility of the results. To avoid permission issues with Docker we will run nextflow being super user (sudo)

Running rnaseq pipeline (scrip7.nf)#
sudo nextflow run script7.nf -with-docker

After few minutes you should have run succesfully your first rnaseq pipeline! Time to check your results:

tree -a results
tree -a work
ls results

Let’s have a look to an interesting report that we get summarizing the different steps of the pipeline. For that you may jave to go to Extensions icon in VSCode and install Live Preview. And then:

code results/multiqc_report.html

Then on the top right corner look for the icon Show Preview.

MultiQC report aggregates results from the pipeline steps (from all the softwares that is compatible with - quite many). In this case it shows some general statistics, fastQC results, and the Quantification information.

In order to avoid to add the option -with-docker, you can customize your config file. Let’s create a config file course.config and add the following:

nextflow.config#
docker.enabled = true

Let’s run again the workflow by calling our config file:

Running rnaseq pipeline (scrip7.nf) with course.config#
sudo nextflow run script7.nf -c course.config

You are are welcome to have a look to the file that was downloaded nextflow.config, to see what usually these files contain and realise how important they are to setup the environment for the nextflow runs. The config file nextflow.config must be in the same folder where you are running your main script (main.nf) otherwise you need to specify the location with -c