The complete Test on Linux 64 bit and Windows 10 TPP 5.2 Docker CLI on LinuxSystem and software - Docker version 1.13.1, build accfe55-unsupported - Docker image: docker.io/spctools/tpp ID ebf55696681a - running on Fedora 29, Intel® Core™ i7-7700HQ CPU @ 2.80GHz, 16 Gb RAM
Data - Reference *data set* from ProteomeXchange repository, identifier PXD001819 (Journal of Proteomics 132 (2016) 51–62), Orbitrap Velos data of human standard proteins (UPS Sigma) spiked into yeast background. - *Conversion of raw data* to .mzML profile, then to .mfg centroid data using docker run -it --privileged=true -v /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast/mzML_profile/:/data chambm/pwiz-skyline-i-agree-to-the-vendor-licenses wine msconvert --mgf /data/*.mzML - *Target data base* ups_human_yeast.fasta, composed of : Uniprot *Homo sapiens* + Uniprot *Saccharomyces cervisiae* + Sigma UPS proteins sequences Size of .mgf files: ~700 Mb each, 27 files FASTA with 33,183 entries Start Docker # systemctl start docker 1. Command Line Interface (CLI) - *Start TPP docker image* with *mounted data directory*: docker run -it --privileged=true -v /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast_centroided/:/data spctools/tpp bash Run comet PSM search - go to data directory data/mgf and *create comet parameters file*: comet -p change name to comet.params and edit database_name = ups_human_yeast.fasta decoy_search = 1 - Start comet search: time comet *.mgf time real 28m45.053s, ~1min/sample, ~140 Mb/sample Run PeptideProphet - in the directory of comet .pep.xml results run: for i in *.pep.xml; do PeptideProphetParser $i; done ~15 s/sample, ~80 Mb/sample Run ProteinProphet - in the directory of PeptideProphet .pep.xml results run: for i in *.pep.xml; do ProteinProphet $i $i.prot.xml NOGROUPS; done ~3 s/sample, ~7 Mb/sample 2. Petunia GUI session (on Linux 64 bit)System and software - Docker version 1.13.1, build accfe55-unsupported - Docker image: docker.io/spctools/tpp ID ebf55696681a - running on Fedora 29, Intel® Core™ i7-7700HQ CPU @ 2.80GHz, 16 Gb RAM Run TPP image - *Start TPP image:* docker run -dit --user=root -p 10401:10401 -v /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast_centroided/:/data spctools/tpp - *Open GUI* with Firefox at http://localhost:10401/tpp/cgi-bin/tpp_gui.pl GUI and guest:guest login works! However: Problems with accessing the directory structure/ uploading files. TPP 5.2 Docker CLI on WindowsSystem and software - Docker version 18.06.1-ce, build e68fc7a - Docker image: docker.io/spctools/tpp ID ebf55696681a - running on OS Name Microsoft Windows 10 Enterprise, Processor Intel® Core™ i5-3470S CPU @ 2.90GHz, 2901 Mhz, 4 Core(s), 4 Logical Processor(s), 8 Gb RAM Data - *MS/MS data* from Waters Synapt DDA experiments, exported to .mgf (see below). 2 samples with ~400 Mb each. - *FASTA database* composed from EST database and Uniref50 plants, created by cat db_1.fasta db_2.fasta > uni50plantsplus.fasta, 3,557,556 entries. Data set not public yet (sorry). Convert Waters Synapt data to mgf - load msconvert image with vendor libraries: docker run -it -v C:\TPP\data:/data chambm/pwiz-skyline-i-agree-to-the-vendor-licenses bash - dive into data directory and run with LockMass Filter: wine msconvert --filt "lockmassRefiner mz=785.8426" --mgf *.raw - exit image by exit. - *Start TPP* docker session: docker run -it -v C:\TPP\data:/data spctools/tpp bash Comet PSM search - use comet.params from above, just change database_name = uni50plantsplus.fasta - with .mgf files, .fasta sequence database and comet.params in the same directory, run: time comet *.mgf Load spectra killed; change in comet.params: spectrum_batch_size = 1000 Run PeptideProphet - in the directory of comet .pep.xml results run: for i in *.pep.xml; do PeptideProphetParser $i; done Run ProteinProphet - in the directory of PeptideProphet .pep.xml results run: for i in *.pep.xml; do ProteinProphet $i $i.prot.xml NOGROUPS; done Convert prot.xml to HTML/EXCEL - In the directory of .prot.xml data run: for i in *.prot.xml; do /usr/local/tpp/cgi-bin/ProtXMLViewer.pl -file $i -action ExportExcel; done Leave TPP docker session by exit and enjoy your results. Conclusion The TPP docker image ebf55696681a (version 5.2) runs on both, Linux and Windows 10 host, in the CLI mode as expected. Am Freitag, 16. November 2018 16:39:06 UTC+1 schrieb Robert: > > TPP 5.2 Docker on Linux > > # System and software > > - Docker version 1.13.1, build accfe55-unsupported > - Docker image: docker.io/spctools/tpp ID ebf55696681a > - running on Fedora 29, Intel(R) Core(TM) i7-7700HQ CPU @ 2.80GHz, 16 Gb > RAM > > # Data > > - Reference **data set** from ProteomeXchange repository, identifier > PXD001819 (Journal of Proteomics 132 (2016) 51–62), Orbitrap Velos data of > human standard proteins (UPS Sigma) spiked into yeast background. > - **Conversion of raw data** to .mzML profile, then to ``.mfg`` centroid > data using > ``docker run -it --privileged=true -v > /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast/mzML_profile/:/data > chambm/pwiz-skyline-i-agree-to-the-vendor-licenses wine msconvert --mgf > /data/*.mzML`` > - **Target data base** ``ups_human_yeast.fasta``, composed of : Uniprot > _Homo sapiens_ + Uniprot _Saccharomyces cervisiae_ + Sigma UPS proteins > sequences > Size of .mgf files: ~700 Mb each, 27 files > FASTA with 33183 entries > > # Start Docker > ``# systemctl start docker`` > > # 1. Command Line Interface (CLI) > - **Start TPP docker image** with **mounted data directory**: > ``docker run -it --privileged=true -v > /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast_centroided/:/data > spctools/tpp bash `` > > ## Run comet PSM search > > - go to data directory data/mgf and **create comet parameters file**: > ``comet -p`` > change name to comet.params and edit > ``database_name = ups_human_yeast.fasta > decoy_search = 1`` > - Start comet search: > ``time comet *.mgf`` > time real 28m45.053s, ~1min/sample, ~140 Mb/sample > > ## Run PeptideProphet > > - in the directory of comet .pep.xml results run: > ``for i in *.pep.xml; do PeptideProphetParser $i; done`` > ~15 s/sample, ~80 Mb/sample > > ## Run ProteinProphet > > - in the directory of PeptideProphet .pep.xml results run: > ``for i in *.pep.xml; do ProteinProphet $i $i.prot.xml NOGROUPS; done`` > ~3 s/sample, ~7 Mb/sample > > ## Convert to HTML/EXCEL > > ``pepxml2html`` not found. > Leaving TPP docker session by ``exit``. > - Conversion with own script < > https://bitbucket.org/lababi/protyquant/src/06deaeb70a09b8121ce0adc1d7d6da389afe7175/python-scripts/protxml_to_tsv.py?at=master>: > > > ``for i in *.prot.xml; do python3 protxml_to_tsv.py $i; done`` > > # 2. Petunia GUI session > > - **Start TPP image:** > ``docker run -dit --user=root -p 10401:10401 -v > /home/robertwinkler/dataspace/nextcloud/DATA/UPS48_yeast_centroided/:/data > spctools/tpp`` > - **Open GUI** with Firefox at < > http://localhost:10401/tpp/cgi-bin/tpp_gui.pl> > GUI and guest:guest login works! > However: Problems with accessing the directory structure/ uploading files. > > > > Am Montag, 22. Oktober 2018 14:49:30 UTC+2 schrieb Robert: >> >> Dear TPP friends, >> >> has anyone tried (or even solved) to run the TPP/Petunia GUI from a >> Docker container/image? >> >> Best, Robert >> > -- You received this message because you are subscribed to the Google Groups "spctools-discuss" group. To unsubscribe from this group and stop receiving emails from it, send an email to [email protected]. To post to this group, send email to [email protected]. Visit this group at https://groups.google.com/group/spctools-discuss. For more options, visit https://groups.google.com/d/optout.
