Skip to content

princelab/mspire

Folders and files

NameName
Last commit message
Last commit date

Latest commit

 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Repository files navigation

mspire

Gem Version Build Status Dependency Status Code Climate

Mspire is a full featured library for working with mass spectrometry data, particularly proteomic, metabolomic and lipidomic data sets. It aims to be fast, robust, and beautiful.

Cite

Prince JT, Marcotte EM. mspire: mass spectrometry proteomics in Ruby. Bioinformatics. 2008 Dec 1;24(23):2796-7. (pubmed)

Features

mzml

  • Reading and writing
  • True random access to spectra or chromatograms
  • Complete object model with implicit object link resolution (even with random access)
  • Simplified creation of and full support for CV params and referenceable param groups

imzml

Mspire is the only commandline converter from mzml into imzml (also see imzMLConverter)

  • handles both processed and continuous modes
  • gracefully handles SIM data

Other Feature Highlights

  • isotope distribution prediction: uses fastest method known (FFT convolution)
  • protein digestion: Support for 32 enzymes/variants by name
  • pepxml: full object model and complete write support
  • fasta files: complete programmatic access to description lines (via bio-ruby)
  • peak lists: merging/summing and splitting algorithms
  • obo: ontology hash access
  • molecular formulas: can do arithmetic with formulas
  • calculates q-values

Examples

mzml_file = "yourfile.mzML"

mzml

See Mspire::Mzml, Mspire::CV::Paramable, Mspire::Mzml::Spectrum and other objects associated with Mzml files.

reading

require 'mspire/mzml'

# get the intensity of the highest peak from each spectrum
intensities = Mspire::Mzml.foreach(mzml_file).map do |spectrum|
  spectrum.intensities.max
end

# open the file for other operations
Mspire::Mzml.open(mzml_file) do |mzml|
  # read each spectra
  mzml.each do |spectrum|
    # do something with each spectrum ...
  end

  # or random access by index or id (even if file wasn't indexed)
  spectrum = mzml[0]
  spectrum = mzml["controllerType=0 controllerNumber=1 scan=2"]

  # some things to do with a spectrum
  spectrum.mzs
  spectrum.intensities

  # first 5 peaks
  spectrum.peaks[0,5].each do |mz, intensity|
    puts "#{mz} #{intensity}"
  end

  # true if key exists and no value, the value if present, or false
  if spectrum.fetch_by_acc('MS:1000128')
    puts "this is a profile spectrum!"
  end

  if spectrum.ms_level == 2
    low_mz = spectrum.scan_list.first.scan_windows.first.fetch_by_acc("MS:1000501").to_i
    puts "begin scan at #{low_mz} m/z"
  end
end

normalize spectra and write new mzML

See Mspire::Mzml for complete example building all objects from scratch.

require 'mspire/mzml'

Mspire::Mzml.open(mzml_file) do |mzml|

  # MS:1000584 -> an mzML file
  mzml.file_description.source_files << Mspire::Mzml::SourceFile[mzml_file].describe!('MS:1000584')
  mspire = Mspire::Mzml::Software.new
  mzml.software_list.push(mspire).uniq_by(&:id)
  normalize_processing = Mspire::Mzml::DataProcessing.new("ms1_normalization") do |dp|
    # 'MS:1001484' -> intensity normalization 
    dp.processing_methods << Mspire::Mzml::ProcessingMethod.new(mspire).describe!('MS:1001484')
  end

  mzml.data_processing_list << normalize_processing

  spectra = mzml.map do |spectrum|
    normalizer = 100.0 / spectrum.intensities.max
    spectrum.intensities.map! {|i| i * normalizer }
    spectrum
  end
  mzml.run.spectrum_list = Mspire::Mzml::SpectrumList.new(normalize_processing, spectra)
  mzml.write(outfile)
end

Masses

# very high precision NIST masses
aa_to_mass = Mspire::Mass::AA::MONO # a hash with residue masses
aa_to_mass['A'] # or access by symbol - Alanine

# elements
Mspire::Mass::Element::MONO[:C] # carbon
Mspire::Mass::Subatomic::MONO[:electron]

Isotopes and molecular formulas

Uses Richard Neese biological isotope ratios by default. All other isotope information culled from NIST.

require 'mspire/isotope'
isotopes = Mspire::Isotope::ISOTOPES  # 288 isotopes
# same as Mspire::Isotope::Neese::ISOTOPES ; compare to Mspire::Isotope::NIST::ISOTOPES
hydrogen_isotopes = isotopes.select {|iso| iso.element == :H }

c12 = Mspire::Isotope::BY_ELEMENT[:C].first
c12.atomic_number # also: mass_number atomic_mass relative_abundance average_mass
c12.mono   # => true (this is the monoisotopic isotope)

require 'mspire/molecular_formula'  # requires fftw gem
propane = Mspire::MolecularFormula['C3H8']
butane = propane + Mspire::MolecularFormula['CH2']
puts butane  # => C4H10

require 'mspire/isotope/distribution'  # requires fftw gem
puts butane.isotope_distribution  # :total, :max, :first as arg to normalize

Digestion

require 'mspire/digester'
trypsin = Mspire::Digester[:trypsin].
trypsin.digest("AACCKDDEERFFKPGG") # => ["AACCK", "DDEER", "FFKPGG"]

TODO

  • write the mzml index onto a file (along with correct SHA-1)
  • implement spectrum unpack into an nmatrix or narray
  • do a proper copy over of meta-data from mzml into imzml
  • consider implementing params as a hash and formalizing more complete implementation agnostic params api

Acronym

Mass SPectrometry In Ruby. Mspire originally stood for Mass Spectrometry Proteomics In Ruby but the library has since proven useful for all kinds of mass spectrometry projects, hence the more inclusive form. The e was originally included for aesthetic reasons, but it also provides the user/developer the ability to attach whatever Exclamation or Epithet they choose to the acronym (the best ones will begin with e of course).

Copyright

MIT license. See LICENSE for details.