ELIXIR QC proteomics pipeline

Global System Requirements:

- HPC Cluster running CentOS 7 x64. Do not install it in a desktop PC.
- Singularity container >= 2.6.1.
- Nextflow >= 20.01.0.5264.
- At least 3GB free disk space.  

Installation:

1. git clone https://github.com/proteomicsunitcrg/qcloud2-pipeline.git, checkout "local" branch and chmod -R 770 at the pipeline root folder.
2. Set up Nextflow params.config file:

params {
    qconfig          = "$baseDir/qc.config"
    zipfiles         = "$baseDir/incoming/*.zip"
    out_folder       = "output"
    fasta_tab        = "$baseDir/fasta.tsv"
    watch            = "YES"
    email            = "roger.olivella@crg.eu"
}

Where:

• qconfig: path to qc.config file. Do not change it.
• zipfiles: incoming folder where the RAW files (zipped) will be put to be processed by the pipeline.
• out_folder: folder where the output QC metrics files are going to be stored.
• fasta_tab: file with used FASTA databases. Do not change it.
• watch: "YES" if you want the pipeline to be automatically started when a file is moved to zipfiles folder.
• email: notification email.

3. Set up nextflow.config file: modify this file according to your HPC Cluster queues name/s and the memory and CPUs available.

Usage:

1. Run the pipeline in background mode: nextflow run -bg elixir_proteomics_QC.nf > elixir_qc.log. The first time you run this command Nextflow will automatically pull the last pipeline container version labeled as biocorecrg/qcloud:2.2 (1.5GB aprox.), ThermoRawFileParser container biocorecrg-thermorawparser-0.1.img (438M) that converts RAW files to mzML format and mzQC converter biocorecrg-mzqc_conv-0.1.img (242M), all three files stored at $baseDir/image.
2. Once all images have been pulled, you can copy any RAW file to the incoming folder. Its filename must be in a specific format to be successfully processed by the pipeline:
   • All files must be zipped with the same name as the RAW file.
   • Its name must follow the following convention: QC02_checksum.raw.zip, where:
     • QC02, wich is an internal code to identify complex human samples (HeLa).
     • Checksum: the md5sum of the RAW file (not the zipped one).
     • For instance: QC02_c010cb81200806e9113919213772aaa9.raw.zip

Output files:

Once the pipeline has been processed all the files moved to $baseDir/incomingfolder, it'll create a folder with the same name as the RAW files in the output folder defined at params.config. Inside this folder you'll find:

• mzQC: the HUPO-PSI standard format file for QC metrics.
• mzID: the HUPO-PSI standard format file for protein identification.
• featureXML: OpenMS features.
• qcML: OpenMS QC metrics.
• mzML: the HUPO-PSI standard format file for mass spec raw files.

Credits (specifically for the bioinformatics part):

• QCloud Pipeline: Luca Cozzuto and Roger Olivella.
• ThermoFileRawParser: Niels Hulstaert (https://github.com/compomics/ThermoRawFileParser#thermorawfileparser). ).

Credits (for the entire project):

Mathias Walzer, Juan Antonio Vizcaíno and Eduard Sabidó.