Using mProphet at ISB
From SPCTools
| Revision as of 01:23, 26 August 2011 Tfarrah (Talk | contribs) ← Previous diff |
Current revision Tfarrah (Talk | contribs) |
||
| Line 1: | Line 1: | ||
| mQuest and mProphet are a software suite for the analysis of SRM data. See [http://www.mprophet.org] for literature reference, [http://www.mprophet.org/download/Manual.pdf manual], and installation instructions. | mQuest and mProphet are a software suite for the analysis of SRM data. See [http://www.mprophet.org] for literature reference, [http://www.mprophet.org/download/Manual.pdf manual], and installation instructions. | ||
| - | To run mQuest, you need a tab-separated transition file with the following columns, only three of which are required: | + | As of March 2013, the most functional versions of the software are those modified by Terry Farrah. They include some bug fixes implemented at ISB. These instructions show you how to use these versions. |
| + | |||
| + | |||
| + | ==Set up transition list== | ||
| + | |||
| + | To run mQuest, you need a tab-separated transition file with the following columns, only three of which are required. Call this file transition_list.tsv: | ||
| *'''Q1 (required)''' | *'''Q1 (required)''' | ||
| *'''Q3 (required)''' | *'''Q3 (required)''' | ||
| Line 8: | Line 13: | ||
| *prec_z | *prec_z | ||
| *frg_type (y, b, etc.) | *frg_type (y, b, etc.) | ||
| - | *frg_nr (3, 7, etc.) | + | *frg_nr (3, 7, etc.) (The Excel formula MID() is useful for parsing y12 into y and 12) |
| *frg_z | *frg_z | ||
| *Isotype (heavy or light) | *Isotype (heavy or light) | ||
| Line 16: | Line 21: | ||
| (See p. 6 of [http://www.mprophet.org/download/Manual.pdf the manual] for other columns you can provide.) | (See p. 6 of [http://www.mprophet.org/download/Manual.pdf the manual] for other columns you can provide.) | ||
| - | Put transition file into same directory as your data (mzML or mzXML format). | + | Put transition file into the same directory as your data (mzML or mzXML format). Don't use symlinks. |
| - | Log onto regis and go to that directory. | + | Log onto regis and go to that directory. Then ... |
| - | Then: | + | ==Set up environment variables== |
| - | ==mzXML Files== | + | If you use the bash shell: |
| - | '''/proteomics/sw/mquest/bin/mInteract.pl -force -mmap -mmap_machine QTRAP -cycle 2.0 -mquest ./ -workflow LABEL_FREE''' | + | '''MPRO_DIR_OLD=/proteomics/sw/mquest.x.x.x/bin/''' |
| - | -cycle is cycle time; used for QTRAP mzXML files only. | + | (If you use C-shell or similar, precede each of these commands with the word <i>set</i>.) |
| - | -force processes all files, even those that have already been processed. | + | Set your machine type (TSQ, QTRAP, or QQQ): |
| - | Other options for mmap_machine: TSQ, QQQ. | + | '''MACHINE=TSQ''' |
| - | Other options for workflow: SPIKE_IN. Type /proteomics/sw/mquest/bin/mInteract.pl -help or -manual for more info. | + | If your data is label-free: |
| - | ==mzML Files== | + | '''MPRO_DEF=label_free.def''' |
| - | mInteract doesn't yet work for mzML files, so you have to run mMap and mQuest separately. | + | |
| - | '''/proteomics/sw/mquest/bin/mMap2.pl -force -machine=QTRAP -cycle=2.0 -strict ./ -mz "*.mzML" -trans transitions.tsv''' | + | If your data is spiked with heavy-labelled peptides: |
| - | -cycle is cycle time; used for QTRAP mzXML files only. | + | '''MPRO_DEF=spike-in.def''' |
| - | -force processes all files, even those that have already been processed. | + | Now, copy the appropriate params file to your directory: |
| - | -mzxml "*mzML" is a specification of the spectrum files to use; it is not needed if you have mzXML files and want to process all of them. | + | '''cp /proteomics/sw/mquest.x.x.x/conf/$MPRO_DEF .''' |
| - | -machine -- for mzML it doesn't matter what you put here. | + | ==Run mMap== |
| + | |||
| + | '''$MPRO_DIR_OLD/mInteract_TF.pl -mMap -input ./ -trans transition_list.tsv -def $MPRO_DEF -mmap_machine $MACHINE -v -force -strict | ||
| + | |||
| + | ==Run mQuest== | ||
| + | |||
| + | '''$MPRO_DIR_OLD/mInteract_TF.pl -mQuest -input ./ -def $MPRO_DEF -v -force''' | ||
| + | |||
| + | -force processes all files, even those that have already been processed. | ||
| - | '''/proteomics/sw/mquest/bin/mQuest.pl -workflow LABEL_FREE ./ ''' | + | ==Run mProphet== |
| + | To run mProphet, you need to have measured decoy peptides. See mProphet documentation for further information on this topic. Add -mprophet to the mInteract command line to run mProphet. There are bugs and results are produced only for some datasets. During 2011 Terry Farrah was in communication with the authors of the software to get the bugs fixed, but the authors never addressed all our concerns, and it appears that the version of the software that is downloadable from the web still has these bugs. | ||
| - | -workflow: see above | ||
| - | ==mProphet== | + | Type /proteomics/sw/mquest.2.0.3/mQuest/bin/mInteract.pl -help or -manual for more info. |
| - | To run mProphet, you need to have measured decoy peptides. See mProphet documentation for further information on this topic. Add -mprophet to the mInteract command line to run mProphet. | + | |
Current revision
mQuest and mProphet are a software suite for the analysis of SRM data. See [1] for literature reference, manual, and installation instructions.
As of March 2013, the most functional versions of the software are those modified by Terry Farrah. They include some bug fixes implemented at ISB. These instructions show you how to use these versions.
Contents |
Set up transition list
To run mQuest, you need a tab-separated transition file with the following columns, only three of which are required. Call this file transition_list.tsv:
- Q1 (required)
- Q3 (required)
- transition_group_id (required) -- a unique id for the transitions of one precursor
- stripped_sequence
- prec_z
- frg_type (y, b, etc.)
- frg_nr (3, 7, etc.) (The Excel formula MID() is useful for parsing y12 into y and 12)
- frg_z
- Isotype (heavy or light)
- protein_name
- CE
(See p. 6 of the manual for other columns you can provide.)
Put transition file into the same directory as your data (mzML or mzXML format). Don't use symlinks.
Log onto regis and go to that directory. Then ...
Set up environment variables
If you use the bash shell:
MPRO_DIR_OLD=/proteomics/sw/mquest.x.x.x/bin/
(If you use C-shell or similar, precede each of these commands with the word set.)
Set your machine type (TSQ, QTRAP, or QQQ):
MACHINE=TSQ
If your data is label-free:
MPRO_DEF=label_free.def
If your data is spiked with heavy-labelled peptides:
MPRO_DEF=spike-in.def
Now, copy the appropriate params file to your directory:
cp /proteomics/sw/mquest.x.x.x/conf/$MPRO_DEF .
Run mMap
$MPRO_DIR_OLD/mInteract_TF.pl -mMap -input ./ -trans transition_list.tsv -def $MPRO_DEF -mmap_machine $MACHINE -v -force -strict
Run mQuest
$MPRO_DIR_OLD/mInteract_TF.pl -mQuest -input ./ -def $MPRO_DEF -v -force
-force processes all files, even those that have already been processed.
Run mProphet
To run mProphet, you need to have measured decoy peptides. See mProphet documentation for further information on this topic. Add -mprophet to the mInteract command line to run mProphet. There are bugs and results are produced only for some datasets. During 2011 Terry Farrah was in communication with the authors of the software to get the bugs fixed, but the authors never addressed all our concerns, and it appears that the version of the software that is downloadable from the web still has these bugs.
Type /proteomics/sw/mquest.2.0.3/mQuest/bin/mInteract.pl -help or -manual for more info.
