dtk

Description:

Python library of PSM analysis tools. Given PSMs and MS2 spectra, supports modules for plotting DRIP PSMs' decoded Viterbi paths and generation of Lorikeet files for interative, in-browser analysis. Within the python interactive shell, supports instantation of PSMs, real-time Viterbi decoding of DRIP PSMs, and plotting of a DRIP PSM's decoded Viterbi path.

Modules:

plot_psms() - given PSM file, decode and plot all DRIP PSMs
Inputs:
1. PSM file in DRIP's output format (required)
2. .ms2 file of observed spectra (required)
3. HTML file to write list of output figures to (if not specified, defaults to 'currPsms.html')
4. Boolean denoting whether high-resolution MS2 spectrum (if unspecified, defaults to False)
5. Used DRIP learned means, output of dripTrain (if unspecified, defaults to '')
6. Used DRIP learned covars, output of dripTrain (if unspecified, defaults to '')
7. Modifications specification (if unspecified, defaults to 'C+57.0214')
8. N-terminal modifications specification (if unspecified, defaults to '')
9. C-terminal modifications specification (if unspecified, defaults to '')
10. Boolean denoting whether to filter the precursor MS1 peak (if unspecified, defaults to False)
11. If high-resolution MS2 spectrum, m/z value to fit 99.9% of theoretical Gaussian mass within (if unspecified, defaults to 0.05)
Output:
- Figure for each PSM in input file of DRIP decoded Viterbi path, listed in the specified HTML file (3rd input)
gen_lorikeet() - generate Lorikeet files for input PSMs. The Lorikeet package is assumed to be unzipped in the run directory
Input:
1. Tab-delimited PSM file with user-specified fields (required)
2. .ms2 file of observed spectra (required)
3. directory to write output Lorikeet HTML files to (required)
4. HTML file to write list of output Lorikeet files to (if not specified, defaults to 'currPsms.html')
5. Modifications specification (if unspecified, defaults to 'C+57.0214')
6. N-terminal modifications specification (if unspecified, defaults to '')
7. C-terminal modifications specification (if unspecified, defaults to '')
8. Header field denoting scan ID number (if unspecified, defaults to 'scan')
9. Header field denoting peptide sequence (if unspecified, defaults to 'sequence')
10. Header field denoting PSM charge (if unspecified, defaults to 'charge')
11. Optional header field denoting score (if unspecified, defaults to 'score')
12. Header field denoting variable mod sequence if variable mods were specified during the search, following DRIP's output specification (if unspecified, defaults to '')
Output:
- Lorikeet file for each PSM in input file, listed in the specified HTML file (4th input)
percolatorPsms_gen_lorikeet() - generate Lorikeet files specifically for Percolator output PSMs (variable mods not supported since this information is not typically output by Percolator). The Lorikeet package is assumed to be unzipped in the run directory
Input:
1. Tab-delimited PSM file with user-specified fields (required)
2. .ms2 file of observed spectra (required)
3. directory to write output Lorikeet HTML files to (required)
4. Boolean denoting whether file was generated by crux percolator (if not specified, defaults to False)
5. HTML file to write list of output Lorikeet files to (if not specified, defaults to 'currPsms.html')
6. Modifications specification (if unspecified, defaults to 'C+57.0214')
7. N-terminal modifications specification (if unspecified, defaults to '')
8. C-terminal modifications specification (if unspecified, defaults to '')
Output:
- Lorikeet file for each PSM in input file, listed in the specified HTML file (5th input)
load_spectra() – Load spectra from .ms2 file.
Input:
- .ms2 file
Output:
- Dictionary whose keys are scan numbers and elements are spectrum objects with relevant attributes .spectrum_id (scan ID number), .intensity (list of intensity values), and .mz (list of m/z values)
psm() – instantiate PSM object.
Inputs:
1. Peptide sequence (required)
2. Observed spectrum (accessed via dictionary returned by load_spectra) (required)
3. PSM charge (if unspecified, defaults to 2)
4. Boolean denoting whether high-resolution MS2 spectrum (if unspecified, defaults to False)
5. Used DRIP learned means, output of dripTrain (if unspecified, defaults to '')
6. Used DRIP learned covars, output of dripTrain (if unspecified, defaults to '')
7. Modifications specification (if unspecified, defaults to 'C+57.0214')
8. N-terminal modifications specification (if unspecified, defaults to '')
9. C-terminal modifications specification (if unspecified, defaults to '')
10. Boolean denoting whether to filter the precursor MS1 peak (if unspecified, defaults to False)
11. If high-resolution MS2 spectrum, m/z value to fit 99.9% of theoretical Gaussian mass within (if unspecified, defaults to 0.05)
Output:
- decoded DRIP PSM object with relevant attributes .peptide (peptide sequence), .spectrum (PSM observed spectrum object), .insertion_sequence (decoded sequence of Booleans denoting whether the ith peak in the observed spectrum is an insertion or not), .used_ions (decoded sequence of non-deleted theoretical peaks)
plot_drip_viterbi() – for p = psm(), p.plot_drip_viterbi(plotName) plots decoded sequence of DRIP scored b- and y-ions to figure plotName.

Example usage:

Step-by-step examples of usage are given here. In what follows, lines beginning with >>> are run within python's shell. To run the following examples, import the library in the python shell using


>>> import dtk

DRIP decode and plot PSMs


>>> psms = 'dripSearch-test-output-beam75.txt'

>>> spectra = 'data/test.ms2'

>>> dtk.plot_psms(psms, spectra, 'currPsms.html')

DRIP decode and plot high-res MS2 PSMs with variable mods


>>> psms = 'dripSearch-malariaTestVarmods-output.txt'

>>> ms2="data/malariaTest.ms2"

>>> mods = '3M+15.9949,C+57.0214,K+229.16293'

>>> nterm_mods = 'X+229.16293'

>>> cterm_mods = ''

>>> dtk.plot_psms(psms, ms2, 'currPsms.html',
              True, '', '',
              mods, nterm_mods)

Generate Lorikeet files for general PSM file with variable mods


>>> ms2="data/malariaTest.ms2"

>>> mods = '3M+15.9949,C+57.0214,K+229.16293'

>>> nterm_mods = 'X+229.16293'

>>> cterm_mods = ''

>>> scanField = 'scan'

>>> chargeField = 'charge'

>>> peptideField = 'sequence'

>>> scoreField = 'percolator score'

>>> psmFile='data/lorikeetData/cruxPercolatorMalariaVarModsTestOutput.txt'

>>> dtk.gen_lorikeet(psmFile, ms2,'genLorikeetPlasmoCruxPlots','genLorikeetPlasmoCruxPlots.html', mods, nterm_mods, cterm_mods, scanField, peptideField, chargeField, scoreField, 'Var_mod_seq')

Generate Lorikeet files for stand-alone Percolator output


>>> ms2="data/test.ms2"

>>> mods = 'C+57.0214'

>>> psmFile='data/lorikeetData/percolatorTestOutput.txt'

>>> dtk.percolatorPsms_gen_lorikeet(psmFile, ms2,'lorikeetPlots',False,'currPsms.html',mods)

DRIP Toolkit home