peptide-spectrum-match

This is a helper class for dealing with peptide spectrum matches. It is essentially a template for an object with a few useful functions thrown in.

This module is still a work in progress. The idea is for this class to be a standard way of dealing with PSMs from X! tandem, mzIdentML, pepXML, and csv files. Anything is subject to change while all the various parsers are written, but things should be in a fairly stable state from here on out.

Rather than try to choose which score should be reported as the score and which as the expect value, I am adding in fields for all the various types of scores (hyperscore, xcorr, etc.). Not all of them are added in yet and the example below is out of date. Don't build anything that depends on the structure of this class yet.

Usage

Quickstart

var peptideSpectrumMatch = require('peptide-spectrum-match');

let psm = {
    'sequence': 'ELVISLIVESK',
    'sequence_pre': 'R',
    'sequence_post': 'L',
    'missed_cleavages': 0,
    'protein': 'Unknown protein',
    'charge': 2,
    'retention_time': 5.0,
    'precursor_mass': 1228.73,
    'mass_err': 0.0022,
    'theoretical_mass': 1228.7278,
    'modifications': [{'residue':'S', 'position':4, 'mass':23.1}],
    'filename': 'spectra.mgf',
    'scan_title': '3.3.1',
    'scan_id': '3',
    'score': 78.6,
    'expect': 3e-10,
    'is_decoy': false,
    'rank': 1,
    'search_engine': 'tandem'
  };

var psmObj = new peptideSpectrumMatch(psm);

If you want to save your PSMs as a csv, there are a couple helper functions to do so.

getHeader(delimiter=",")

This function will return a header string for a csv file with the specified delimiter. The returned string does not have a newline character at the end of it.

getDelimited(delimiter=",")

This function will return all the fields of the peptide-spectrum-match object as a delimited string. The values will be in the same order as what is indicated by the header. Once again there is no newline character at the end of the string.

The fields of each peptide-spectrum-match object are:

• sequence which is the sequence of the matched peptide
• sequence_pre which is the residue immediately before the sequence
• sequence_post which is the residue immediately after the sequence
• missed_cleavages is the reported number of missed cleavages
• protein is the FASTA entry against which this spectrum was matched
• charge is the charge of the spectrum
• retention_time is the retention time in minutes
• precursor_mass is the observed precursor neutral mass
• mass_err is the deviation of the observed mass from the theoretical. It is the precursor_mass - theoretical_mass
• theoretical_mass is the calculated neutral mass of the peptide
• modifications is a list of objects which specify the residue, position (0-based), and mass of the modification
• filename is the filename of the file from which the spectrum came
• scan_title is the title of the spectrum
• scan_id is the ID of the spectrum
• score is the score assigned by the search engine
• expect is the expectation value assigned by the search engine
• is_decoy is a boolean specifying whether or not the match is to a decoy sequence
• rank is the rank of the match
• search_engine is the name of the engine used to make the match

Tests

You can run npm test to run the tests after installing the development dependencies.

Future functionality

The name may be added in to the entries of the modifications array once a method is developed for handling cases where the modification doesn't come with a name.

License

This software is released under the MIT license.

Support this project!