Repository files navigation

                                              _____________                                              < mspipeline1 >                                              -------------                                                           \                              ___......__             _     \                         _.-'           ~-_       _.=a~~-_   --=====-.-.-_----------~   .--.       _   -.__.-~ ( ___===>               '''--...__  (    \ \\\ { )       _.-~                                 =_ ~_  \\-~~~//~~~~-=-~                                      |-=-~_ \\   \\                                              |_/   =. )   ~}                                             |}      ||                                                 //       ||                                               _//        {{                                            '='~'          \\_    =                                                     ~~'

If you want to use fragpipe using the command line interface, then this is the tool for you.

This pipeline takes 1) a list of .d files and 2) a list of fasta-amino acid files and outputs sane protein calls with abundances. It uses philosopher database and fragpipe to do the job. The snakemake pipeline maintains a nice output file tree.

Because it makes sure that all outputs are updated when you change input-parameters. It also yells at you if something fails, and hopefully makes it a bit easier to find the error.

1. Prerequisites:

• Preferably a HPC system, or a beefy local workstation.
• An conda package manager on that system. (We recommendminiforge)

2. Clone this repo on the HPC/workstation where you want to work.

   git clone https://github.com/cmkobel/mspipeline1.git && cd mspipeline1

3. If you don't already have an environment with snakemake and mamba installed, use the following command to install a "snakemake" environment with the bundled environment file:

   conda env create -f environment.yaml -n mspipeline1

   This environment can then be activated by typingconda activate mspipeline1

4. If needed, tweak the profiles/slurm/ configuration so that it matches your execution environment. There is a profile for local execution without a job management system (profiles/local/) as well as a few profiles for different HPC environments like PBS and SLURM.

The file config_template.yaml contains all the parameters needed to run this pipeline. You should change the parameters to reflect your sample batch.

Because nesvilab do not make their executables immediately publicly available, you need to tell the pipeline where to find them on your system. Update addresses for the keysphilosopher_executable,msfragger_jar,ionquant_jar andfragpipe_executable which can be downloadedhere,here,here andhere, respectively.

Currently the pipeline is only tested on the input of .d-files (agilent/bruker): Create an item in batch_parameters where you define keyd_base which is the base directory where all .d-files reside. Define keydatabase_glob which is a path (or glob) to the fasta-amino acid files that you want to include in the target protein database.

Define items under thesamples key which link sample names to the .d-files.

Lastly, set thebatch key to point at the batch that you want to run.

Finally, run the pipeline in your command line with:

$ snakemake --profile profiles/slurm/

Below is visualization of the workflow graph:

This pipeline might involve an R-markdown performing trivial QC.Also, a test data set that accelerates the development cycle. 🚴‍♀️