PROseq_alignment

Modified from https://github.com/JAJ256/PROseq_alignment.sh

This is a pipeline script for aligning paired-end PRO-seq data that has cells of a different species spiked in for normalization, and uses some combination of random UMI sequences on the ligation end of either the 5' or 3' adapter, or both.

Development and maintenance

Implemented in nextflow by Elin Axelsson-Ekker. She also maintains the pipeline. Please contact her for bug reports, feature requests etc. ([email protected]).

Files and Data needed

• Unaligned sequencing data, as fastq (.fastq/.fq/.fastq.gz or .fq.gz) files. MUST be Paired-End

Output files

The pipeline results in the following folders with output files:

• logs_and_QC
  • align: alignment logs for spikein and experiment alignment
  • dedup: deduplication logs for spikein and experiment alignment
  • fastp: trimming logs for each sample and read
  • fastqc: fastqc output for each sample and read
  • rRNA: bowtie2 logs for rRNA filtering step
  • info: "infoTable.tsv" summary of output
• bams
  • dedup: bam from alignment with experimental genome (deduplication)
• bw
  • unnorm: un-normalized bw files

Organism/annotation

At the moment the pipeline works only for s.pombe with spikein from s. cerevisiae. Contact me if you need other organisms.

Recommended setup

-1. Add Hinkskalle to singularity (one time only)

See How to add Hinkskalle to singularity

0. Start a new tmux session or attach to an existing (optional but useful)

See e.g. https://tmuxcheatsheet.com/ and section Short intro to tmux

1. Clone the repo

From e.g. your user folder in the lab folder do:

git clone https://ngs.vbcf.ac.at/repo/berger_pipelines/PROalign<>.git

you might be asked for user and password, use the ones from your forskalle account

2. Create a folder with the inputs files (or links to the input files)

e.g:

mkdir -p fastq
ln -s /paths/to/files/*.fastq fastq/

As long as you use links and do not actually copy the files, this folder can be created in your user folder on the lab folder, e.g. in the folder that you cloned or in the folder above.

3. Then load the following modules:

NOTE On CBE, you can skip loading singularity

module load nextflow/21.04.1  #or higher if you prefer
module load singularity/3.6.2

NOTE if you are in a tmux session were this was already done, this step can be skipped!

4. If you have not already, then go to the folder.

cd <>

How to run

Make sure you have:

1. completed the setup steps
2. that you are on the cluster with the correct modules loaded
3. moved into the PROalign folder

To use the pipeline on data with UMI on both sides, with the data files ending with R1_001.fastq.gz and R2_001.fastq.gz:

    nextflow run PROalign.nf -profile slurm --files "/path/to/files/*_R{1,2}_001.fastq.gz" \
    --FIVEP_UMI true \
    --THREEP_UMI true \
    -w /path/to/scratch/dir

Here the "/path/to/files/" is the path to the folder you created earlier. The -w sets the path to where the pipeline should be run and where the intermediate files will be stored. It should be on the scratch and not in the lab folder!

NOTE that the files path needs to be within citation marks ("") otherwise nextflow will only use one bam file from the folder and not all.

Parameters

Required

• --files: path to input files, see also examples below.
• --FIVEP_UMI: five prime umi? false or true
• --THREEP_UMI: three prime umi? false or true

Examples files parameter

• Fastq alt1:
  • --files "path/to/files/*_{1,2}.fastq"
  • --files "path/to/files/*_{1,2}.fq"
• Fastq alt2: compressed
  • --files "path/to/files/*_{1,2}.fastq.gz"
  • --files "path/to/files/*_{1,2}.fq.gz"

In the above examples /path/to/files/ is the path to the folder with the data. Note the citation marks around the files path!! Fastq files have to have the extension .fq or .fastq. (.fq.gz or .fastq.gz )

If the input data is paired end and in fastq format, then the first and second reads should be indicated in the file names. Usually the name of the files end with _1 and _2 followed by .fq or .fastq (or .fq.gz/.fastq.gz if compressed).

The files from the NGS facility are ending with 'R1_001.fastq.gz' and 'R2_001.fastq.gz'. To use those files set the files flag to:

--files "path/to/files/*_R{1,2}_001.fastq.gz"

With defaults:

• --output: the name of the output folder. [ "results" ]
• --UMI_LEN: [ 6 ]
• --MAPQ: [ 10 ]

More nextflow options:

• to resume add -resume.
• to run in background add -bg

See also https://www.nextflow.io/docs/latest/cli.html for full list of options.

How to add Hinkskalle to singularity

As the pipeline uses containers from the ngs registry (aka "Hinkskalle"), you need to do the following (on the CBE cluster):

    singularity remote add --no-login hinkskalle singularity.ngs.vbcf.ac.at
    singularity remote use hinkskalle

It's enough if you do this once. The settings will then be saved a folder (.singularity) in your home directory and you do not need to worry about it any more. UNLESS you later change the remote registry (singularity remote use). Then you have to re-run the last line before you start the pipeline.

NOTE: If you are using another cluster (e.g. outside of VBC), you need to make sure Singularity (version 3.4 or higher) is i nstalled and loaded BEFORE you add the hinklist registry.

Short intro to tmux

tmux is a very convent tool to use on the cluster. It has many benefits but some of the most obvious are:

1. It lets you set up an environment on the cluster that you can move in an out from as much you want. E.g. if you have a tmux session called 'proseq' you can load the modules needed to run the pipeline (nextflow and singularity) once and then by re-attaching to this session the modules are already loaded.
2. Scripts that are running in the session will continue to run even if you logout from the cluster.

See also e.g. https://en.wikipedia.org/wiki/Tmux/

To start a new session type:

tmux new -s proseq

This will create a tmux session with the name proseq and attach you to it. To detach from the session press Ctrl + b and then d To attach to an existing session type

  tmux a -t proseq

This will put you into the existing session called proseq. Note that the session needs to be created before you can re-attach to it.