QuantMS Hands-On for PXD04621

QuantMS Hands-On for PXD04621#

Open GitHub codespace#

Use the following link to open a GitHub codespace with most of the required software installed:

⚠️ If you do it manually, make sure to select the bigger machine with 4 cores and 16GB RAM

Open in Codespace deeplink

Download the data#

Currently the data is stored on a Google Drive folder. It should be downloaded when you run the tutorial in a GitHub codespace automatically.

In case you do not see the mzML files in the data/PXD040621/mzML/ folder, you can manually use gdown to download these files from Google Drive:

conda activate base
pip install gdown
python 0_download_PXD040621_data.py

Run the analysis#

# export NXF_VER=24.10.6
nextflow run bigbio/quantms \
         -revision 1.5.0 \
         -params-file PXD040621_w_contaminants-params.yaml \
         -profile docker \
         -resume

If you run locally on a Mac with Apple Silicion (M-ships), you need to addtionally the arm profile:

# export NXF_VER=24.10.6
nextflow run bigbio/quantms \
         -revision 1.5.0 \
         -params-file PXD040621_w_contaminants-params.yaml \
         -profile docker,arm \
         -resume

Copy files for further analysis#

cp -aL results/PXD040621/proteomicslfq/. data/PXD040621/processed/
# And maybe save the parameters to reproduce the analysis (for 1.4.0 and above):
cp -aL results/PXD040621/pipeline_info/. data/PXD040621/processed/pipeline_info/

Now it is safe to delete the results/PXD040621 and work folder.

rm -r results/PXD040621 work

Clean up unused docker images#

As we are running in a GitHub codespace, we have limited storage. Therefore let’s clean up our docker images store after the analysis is done.

docker images # see all images
docker image prune -a

Free up some more space#

Should not be necessary

# some cache files
rm -r  /.codespaces/bin/cache/bin/linux-x64/
ls /vscode/extensionsCache/
ls /vscode/serverCache/

Check used disk space#

Check the local storage usage (you have maximum of 32GB in a GitHub codespace) in the root folder with the following command:

du -hd 1 /

Run a different analysis#

  • feel free to run the project PXD041301 in quantms if you want an exercise

⚠️ not tested in codespace and probably too big for the codespace storage

Contents