mspire-lipid 0.2.0

data/scratch/mol_methods.txt

@@ -0,0 +1,183 @@
+#<OpenBabel::OBMol:0x00000000b71c90>
+---
+- add
+- reserve_atoms
+- create_atom
+- create_bond
+- create_residue
+- destroy_atom
+- destroy_bond
+- destroy_residue
+- add_atom
+- insert_atom
+- add_bond
+- add_residue
+- new_atom
+- new_bond
+- new_residue
+- delete_atom
+- delete_bond
+- delete_residue
+- begin_modify
+- end_modify
+- get_mod
+- increment_mod
+- decrement_mod
+- get_flags
+- get_title
+- num_atoms
+- num_bonds
+- num_hvy_atoms
+- num_residues
+- num_rotors
+- get_atom
+- get_atom_by_id
+- get_first_atom
+- get_bond_by_id
+- get_bond
+- get_residue
+- get_internal_coord
+- get_torsion
+- get_angle
+- get_formula
+- get_spaced_formula
+- get_mol_wt
+- get_exact_mass
+- get_total_charge
+- get_total_spin_multiplicity
+- get_dimension
+- get_coordinates
+- get_sssr
+- get_lssr
+- automatic_formal_charge
+- automatic_partial_charge
+- set_title
+- set_formula
+- set_energy
+- set_dimension
+- set_total_charge
+- set_total_spin_multiplicity
+- set_internal_coord
+- set_automatic_formal_charge
+- set_automatic_partial_charge
+- set_aromatic_perceived
+- set_sssrperceived
+- set_lssrperceived
+- set_ring_atoms_and_bonds_perceived
+- set_atom_types_perceived
+- set_ring_types_perceived
+- set_chains_perceived
+- set_chirality_perceived
+- set_partial_charges_perceived
+- set_hybridization_perceived
+- set_implicit_valence_perceived
+- set_kekule_perceived
+- set_closure_bonds_perceived
+- set_hydrogens_added
+- set_corrected_for_ph
+- set_aromatic_corrected
+- set_spin_multiplicity_assigned
+- set_flags
+- unset_aromatic_perceived
+- unset_sssrperceived
+- unset_ring_types_perceived
+- unset_partial_charges_perceived
+- unset_implicit_valence_perceived
+- unset_hydrogens_added
+- unset_flag
+- renumber_atoms
+- set_coordinates
+- to_inertial_frame
+- translate
+- rotate
+- kekulize
+- perceive_kekule_bonds
+- new_perceive_kekule_bonds
+- delete_hydrogens
+- delete_non_polar_hydrogens
+- delete_hydrogen
+- add_hydrogens
+- add_polar_hydrogens
+- strip_salts
+- separate
+- get_next_fragment
+- convert_dative_bonds
+- correct_for_ph
+- assign_spin_multiplicity
+- set_is_pattern_structure
+- center
+- set_torsion
+- find_sssr
+- find_lssr
+- find_ring_atoms_and_bonds
+- find_chiral_centers
+- find_children
+- find_largest_fragment
+- contig_frag_list
+- align
+- connect_the_dots
+- perceive_bond_orders
+- find_angles
+- find_torsions
+- get_gtdvector
+- get_givector
+- get_gidvector
+- has_2d
+- has_3d
+- has_non_zero_coords
+- has_aromatic_perceived
+- has_sssrperceived
+- has_lssrperceived
+- has_ring_atoms_and_bonds_perceived
+- has_atom_types_perceived
+- has_ring_types_perceived
+- has_chirality_perceived
+- has_partial_charges_perceived
+- has_hybridization_perceived
+- has_implicit_valence_perceived
+- has_kekule_perceived
+- has_closure_bonds_perceived
+- has_chains_perceived
+- has_hydrogens_added
+- has_aromatic_corrected
+- is_corrected_for_ph
+- has_spin_multiplicity_assigned
+- is_chiral
+- empty
+- num_conformers
+- set_conformers
+- add_conformer
+- set_conformer
+- copy_conformer
+- delete_conformer
+- get_conformer
+- set_energies
+- get_energies
+- get_energy
+- begin_conformer
+- next_conformer
+- get_conformers
+- begin_atoms
+- end_atoms
+- begin_bonds
+- end_bonds
+- begin_residues
+- end_residues
+- begin_atom
+- next_atom
+- begin_bond
+- next_bond
+- begin_residue
+- next_residue
+- begin_internal_coord
+- next_internal_coord
+- clear
+- do_transformations
+- has_data
+- delete_data
+- clone_data
+- data_size
+- get_all_data
+- get_data
+- begin_data
+- end_data

data/scratch/split_molecules.rb

@@ -0,0 +1,93 @@
+#!/usr/bin/env ruby
+
+require 'yaml'
+require 'openbabel'
+include OpenBabel
+
+if ARGV.size == 0
+  puts "smiles and smart"
+  exit
+end
+
+(smiles_string, smart_search) = ARGV
+
+smi_to_svg = OBConversion.new
+smi_to_svg.set_in_and_out_formats("smi", "svg")
+
+obmol = OBMol.new
+
+smi_to_svg.read_string(obmol, smiles_string)
+obmol.add_hydrogens
+
+p obmol.get_formula
+svg_string = smi_to_svg.write_string(obmol).rstrip
+File.write("before.svg", svg_string)
+
+bond_triplets = []
+(1..obmol.num_bonds).each do |i|
+  bond = obmol.get_bond(i)
+  if bond
+    bond_triplets << [bond.get_begin_atom, bond.get_end_atom, bond]
+  end
+end
+
+(1..obmol.num_atoms).each do |i|
+  atom = obmol.get_atom(i)
+  if atom.matches_smarts("C(=O)CC")
+    bond_triplet = bond_triplets.find {|triplet| triplet.include?(atom) }
+    obmol.delete_bond(bond_triplet.last)
+    atom.begin_nbr_atom 
+    atom.next_nbr_atom
+    break
+  end
+end
+
+molecules = obmol.separate
+molecules.each_with_index do |mol,i|
+  puts "mol #{i}"
+  p mol.get_formula
+  p mol.add_hydrogens
+  p mol.get_formula
+  svg_string = smi_to_svg.write_string(mol).rstrip
+  File.write("after_#{i}.svg", svg_string)
+end
+
+
+
+#puts (obmol.methods - Object.new.methods).map(&:to_s).to_yaml
+#p obmol.exact_mass
+
+
+
+
+=begin
+$obConversion->ReadFile($obMol, $filename);
+
+for (1..$obMol->NumAtoms()) {
+    $atom = $obMol->GetAtom($_);
+    # look to see if this atom is a thiophene sulfur atom
+    if ($atom->MatchesSMARTS("[#16D2]([#6D3H1])[#6D3H1]")) {
+	$sulfurIdx = $atom->GetIdx();
+    # see if this atom is one of the carbon atoms bonded to a thiophene sulfur
+    } elsif ($atom->MatchesSMARTS("[#6D3H1]([#16D2][#6D3H1])[#6]") ) {
+	if ($c2Idx == 0) { $c2Idx = $atom->GetIdx(); }
+	else {$c5Idx = $atom->GetIdx(); }
+    }
+}
+
+# Get the actual atom objects -- indexing will change as atoms are added and deleted!
+$sulfurAtom = $obMol->GetAtom($sulfurIdx);
+$c2Atom = $obMol->GetAtom($c2Idx);
+$c5Atom = $obMol->GetAtom($c5Idx);
+
+$obMol->DeleteAtom($sulfurAtom);
+
+$obMol->DeleteHydrogens($c2Atom);
+$obMol->DeleteHydrogens($c5Atom);
+
+$c2Atom->SetAtomicNum(1);
+$c5Atom->SetAtomicNum(1);
+
+$obConversion->WriteFile($obMol, "$filename.mol");
+
+=end

data/script/find_nearest_lipid.rb

@@ -0,0 +1,134 @@
+#!/usr/bin/env ruby
+
+require 'trollop'
+
+require 'bsearch'
+
+require 'mspire/lipid/ion'
+require 'mspire/lipid/modification'
+require 'mspire/lipid_maps'
+
+parser = Trollop::Parser.new do
+  banner "usage: #{File.basename(__FILE__)} <lipidmaps.tsv> <m/z> ..."
+  banner ""
+  opt :top_n, "how many closest ions to print", :default => 3
+  banner ""
+  text "modifications: (at least 1 charged mod is required)"
+  text "will search for i down to 1 mod: "
+  #opt :newmod, "<name>:<formula>:<charge>:<L/G>:<#occurences>  L/G=loss or gain", :type => :string, :multi => true
+  opt :lithium, "Li+ gain", :default => 0
+  opt :sodium, "Na+ gain", :default => 0
+  opt :ammonium, "", :default => 0
+  opt :ammonia_loss, "search for i down to 1 ammonia losses", :default => 0
+  opt :carbon_dioxide_loss, "search for i down to 1 CO2 losses", :default => 0
+  opt :proton_gain, "search for i down to 1 proton additions", :default => 0
+  opt :proton_loss, "search for i down to 1 proton losses", :default => 0
+  opt :water_loss, "if used, *all* mods are also considered with i down to 0 water losses", :default => 0
+  opt :acetate, "search for i down to 0 water losses", :default => 0
+  text ""
+  text "other:"
+  opt :textfile, "a text file with m/z values, one per line", :type => String
+  opt :lower_bound, "use lower bound searching (requires m/z's to be in sorted order)"
+  opt :sort, "sorts the m/z's"
+end
+
+opts = parser.parse(ARGV)
+
+if ARGV.size == 0
+  parser.educate
+  exit
+end
+
+CHARGED_MODS = [:lithium, :sodium, :ammonium, :proton_gain, :proton_loss]
+UNCHARGED_MODS = [:water_loss, :carbon_dioxide_loss, :ammonia_loss]
+
+unless CHARGED_MODS.any? {|key| opts[key] > 0 }
+  puts "*" * 78
+  puts "ArgumentError: need at least one charged mod!"
+  puts "*" * 78
+  parser.educate
+  exit
+end
+
+(lipidmaps, *actual_mzs) = ARGV
+
+if f=opts[:textfile]
+  actual_mzs = IO.readlines(f).map(&:chomp)
+end
+
+actual_mzs.map!(&:to_f)
+
+unless actual_mzs.size > 0
+  STDERR.puts "NO m/z values given!!!"
+  parser.educate
+  exit
+end
+
+$VERBOSE = opts[:verbose]
+
+LipidMod = Mspire::Lipid::Modification
+
+mods = {
+  proton_gain: LipidMod.new(:proton),
+  water_loss: LipidMod.new(:water, :loss => true),
+  lithium: LipidMod.new(:lithium),
+  sodium: LipidMod.new(:sodium),
+  ammonium: LipidMod.new(:ammonium),
+  proton_loss: LipidMod.new(:proton, :loss => true),
+  ammonia_loss: LipidMod.new(:ammonium, :loss => true),
+  carbon_dioxide_loss: LipidMod.new(:carbon_dioxide, :loss => true),
+}
+
+lipids = Mspire::LipidMaps.parse_file(lipidmaps)
+
+ions = []
+lipids.each do |lipid| 
+  CHARGED_MODS.each do |key|
+    if opts[key] > 0
+      opts[key].downto(1) do |num_charge_mod|
+        mods_to_use = [mods[key]] * num_charge_mod
+        UNCHARGED_MODS.each do |umod|
+          opts[umod].downto(0) do |i|
+            ions << Mspire::Lipid::Ion.new(lipid, mods_to_use + ([mods[umod]]*i)) 
+          end
+        end
+      end
+    end
+  end
+end
+
+ions.sort_by!(&:mz)
+
+PPM = 1500
+
+cats = %w(exp_mz rank ppm ppm_abs category lm_id common_name modifications)
+puts cats.join("\t")
+
+actual_mzs.sort! if opts[:sort]
+
+starting_i = 0
+len = ions.size
+
+actual_mzs.each do |exp_mz|
+
+  range = ions.bsearch_range(starting_i...len) do |ion| 
+    part_diff = ((exp_mz - ion.mz)/ion.mz)*1e6
+    if part_diff.abs <= PPM then 0
+    elsif part_diff < PPM then 1
+    else 
+      -1
+    end
+  end
+
+  starting_i = range.begin if opts[:lower_bound]
+  
+  closest = ions[range].sort_by {|ion| [(ion.mz - exp_mz).abs, ion.mz] }
+  row = [exp_mz]
+  closest[0,opts[:top_n]].each_with_index do |ion,i|
+    rank = i + 1
+    ppm = ((exp_mz - ion.mz) / ion.mz) * 1e6
+    lipid = ion.lipid
+    row.push( rank, ppm, ppm.abs, lipid.category, lipid.lm_id, lipid.common_name, ion.modifications.map(&:charged_formula_string).join(", ") )
+  end
+  puts row.join("\t")
+end

data/spec/mspire/lipid/ion_spec.rb

@@ -0,0 +1,96 @@
+require 'spec_helper'
+
+require 'mspire/lipid'
+require 'mspire/lipid/modification'
+require 'mspire/lipid/ion'
+require 'mspire/mass/element'
+require 'mspire/mass/common'
+
+module MSS
+  CO2 = ( %w(C O O).map {|e| Mspire::Mass::Element::MONO_STRING[e] }.reduce(:+) )
+  H2O = Mspire::Mass::Common::MONO_STRING['H2O']
+  PROTON_LOSS = Mspire::Lipid::Modification.new(:proton, :loss => true)
+end
+
+
+describe Mspire::Lipid::Ion do
+
+  before do
+    lipid = Mspire::Lipid.new("LMGP02010009", "PE(16:0/18:1(9Z))", "1-hexadecanoyl-2-(9Z-octadecenoyl)-sn-glycero-3-phosphoethanolamine", 'C39H76NO8P', 717.5308, 'Glycerophospholipids [GP]', 'Glycerophosphoethanolamines [GP02]', 'Diacylglycerophosphoethanolamines [GP0201]', '7850611', 'FHQVHHIBKUMWTI-OTMQOFQLSA-N')
+    proton = Mspire::Lipid::Modification.new(:proton)
+    proton_loss = Mspire::Lipid::Modification.new(:proton, :loss => true)
+    h2o_loss = Mspire::Lipid::Modification.new(:water, :loss => true)
+    @plus1_less_h20 = Mspire::Lipid::Ion.new(lipid, [proton, h2o_loss])
+    @plus2_less_h20 = Mspire::Lipid::Ion.new(lipid, [proton, proton, h2o_loss])
+    @minus1_less_h20 = Mspire::Lipid::Ion.new(lipid, [proton_loss, h2o_loss])
+  end
+
+  it 'calculates the correct m/z' do
+    @plus1_less_h20.mz.should be_within(1e-5).of(700.52751178307)
+    @plus2_less_h20.mz.should be_within(1e-5).of(350.76739412492003)
+    @minus1_less_h20.mz.should be_within(1e-5).of(698.5129588842301)
+  end
+
+  it 'calculates the correct formula' do
+    @plus1_less_h20.formula.to_s.should == "C39H75NO7P"
+    @plus1_less_h20.charge.should == 1
+
+    @plus2_less_h20.formula.to_s.should == "C39H76NO7P"
+    @plus2_less_h20.charge.should == 2
+    @minus1_less_h20.formula.to_s.should == "C39H73NO7P"
+    @minus1_less_h20.charge.should == -1
+  end
+
+  describe 'predicting ms/ms fragments' do
+    describe 'predicting simple fatty acyls' do
+      xit 'hexadecanoic acid' do
+        lipid = Mspire::Lipid.new('LMFA01010001', 'Palmitic acid', 'hexadecanoic acid',	'C16H32O2', 256.24, 'Fatty Acyls [FA]', 'Fatty Acids and Conjugates [FA01]', 'Straight chain fatty acids [FA0101]')
+
+        ion = Mspire::Lipid::Ion.new(lipid, [MSS::PROTON_LOSS])
+        frags = [ion.mz - MSS::H2O]
+        frags << ion.mz - MSS::CO2
+        frags << ion.mz - (MSS::H2O + MSS::CO2)
+
+        mzs = ion.predict_fragment_mzs
+        mzs.should be_an(Array)
+        mzs.sort.should == frags.sort
+      end
+
+      # WORKING THIS GUY OUT!!!!!!
+      xit '11-methyl-9S-hydroxy-13E-hexadecenoic acid' do
+        lipid = Mspire::Lipid.new(nil, 'Made Up', '11-methyl-9S-hydroxy-13E-hexadecenoic acid',	'C17H32O3', 284.23514488492003, 'Fatty Acyls [FA]', 'Fatty Acids and Conjugates [FA01]', 'Branched fatty acids [FA0102]')
+        ion = Mspire::Lipid::Ion.new(lipid, [MSS::PROTON_LOSS])
+
+        # one water loss
+        frags = [ion.mz - MSS::H2O]
+
+        # two water loss
+        frags << (ion.mz - (2*MSS::H2O))
+
+        # CO2 loss
+        frags << (ion.mz - MSS::CO2)
+
+        # CO2 + H20 loss
+        frags << ion.mz - (MSS::H2O + MSS::CO2)
+
+        oh_frag_mass = 140.12011513268
+        frag1 = (ion.mz - oh_frag_mass)
+
+        frags << frag1 - MSS::H2O
+
+        frags << frag1 - MSS::CO2
+
+        frags << frag1 - (MSS::H2O + MSS::CO2)
+
+        mzs = ion.predict_fragment_mzs
+        #mzs.sort.should == frags.sort
+      end
+
+    end
+
+    it 'predicts simple glycerophosopholipids' do
+    end
+
+  end
+
+end