Module: Rubabel::Molecule::Fragmentable

Included in:
Rubabel::Molecule
Defined in:
lib/rubabel/molecule/fragmentable.rb

Constant Summary collapse

RULES =

:sp3c_oxygen_asymmetric_far_sp3, :sp3c_nitrogen_asymmetric_far_sp3,

Set[ :alcohol_to_aldehyde, :peroxy_to_carboxy, :co2_loss, 
  :sp3c_oxygen_double_bond_far_side_sp3, :sp3c_oxygen_double_bond_far_side_sp2, :sp3c_oxygen_double_bond_water_loss, :sp3c_nitrogen_double_bond,
]
CO_RULES =

ADDUCTS = [:lioh, :nh4cl, :nh4oh]

Set[:alcohol_to_aldehyde, :peroxy_to_carboxy, :co2_loss, 
  :sp3c_oxygen_double_bond_water_loss, :sp3c_oxygen_double_bond_far_side_sp2, :sp3c_oxygen_double_bond_far_side_sp3, :sp3c_oxygen_asymmetric_far_sp3
]
DEFAULT_OPTIONS = 
{
  rules: RULES,
  #adduct: nil,
  ph: 7.4,
  # return only the set of unique fragments
  uniq: false, 
}

Instance Method Summary collapse

Instance Method Details

#alcohol_to_aldehyde(carbon, oxygen, carbon_nbrs) ⇒ Object 



101
102
103
104
105
106
107
108
109
110
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 101

def alcohol_to_aldehyde(carbon, oxygen, carbon_nbrs)
  # alcohol becomes a ketone and one R group is released
  frag_sets = carbon_nbrs.select {|atom| atom.type == 'C3' }.map do |_atom|
    frags = feint_double_bond(carbon.get_bond(oxygen)) do |_mol|
      frags = _mol.split(carbon.get_bond(_atom))
      frags.map(&:add_h!)
    end
  end
  allowable_fragment_sets!(frag_sets)
end

#allowable_fragment_sets!(fragment_sets) ⇒ Object

add_h! to self, then selects allowable fragments



38
39
40
41
42
43
44
45
46
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 38

def allowable_fragment_sets!(fragment_sets)
  self.add_h!
  fragment_sets.select do |_frags| 
    putsv "ExMAIN:"
    putsv _frags.inspect
    putsv self.allowable_fragmentation?(_frags)
    self.allowable_fragmentation?(_frags)
  end
end

#allowable_fragmentation?(frags) ⇒ Boolean

molecules and fragments should all have hydrogens added (add_h!) before calling this method

For instance, water loss with double bond formation is not allowable for NCC(O)CC => CCC=C, presumably because of the lone pair and double bond resonance.

Returns:

  • (Boolean) 


33
34
35
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 33

def allowable_fragmentation?(frags)
  self.num_atoms == frags.map(&:num_atoms).reduce(:+)
end

#co2_loss(carbon, oxygen, c3_nbr) ⇒ Object 



112
113
114
115
116
117
118
119
120
121
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 112

def co2_loss(carbon, oxygen, c3_nbr)
  # carboxyl rules ...
  # neutral carbon dioxide loss with anion gain on attaching group
  # (if carbon)
  frags = feint_double_bond(carbon.get_bond(oxygen), oxygen, c3_nbr) do |_mol|
    frags = _mol.split(c3_nbr.get_bond(carbon))
    frags.map(&:add_h!)
  end
  allowable_fragment_sets!([frags])
end

#electrophile_snatches_electrons(carbon, electrophile) ⇒ Object

warning, this method adds_h! to the calling molecule

Raises:

  • (NotImplementedError) 


65
66
67
68
69
70
71
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 65

def electrophile_snatches_electrons(carbon, electrophile)
  self.add_h!
  frags = self.split(carbon.get_bond(electrophile))
  raise NotImplementedError
  # don't check for allowable fragments because it 
  #allowable_fragment_sets!([frag_set])
end

#feint_double_bond(bond, give_e_pair = nil, get_e_pair = nil, &block) ⇒ Object

will turn bond into a double bond, yield the changed molecule, then return the bond to the original state when the block is closed returns whatever the block returned



51
52
53
54
55
56
57
58
59
60
61
62
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 51

def feint_double_bond(bond, give_e_pair=nil, get_e_pair=nil, &block)
  orig = bond.bond_order
  bond.bond_order = 2
  reply = 
    if give_e_pair || get_e_pair
      feint_e_transfer(give_e_pair, get_e_pair, &block)
    else
      block.call(self)
    end
  bond.bond_order = orig
  reply
end

#feint_e_transfer(give_e_pair = nil, get_e_pair = nil, &block) ⇒ Object 



73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 73

def feint_e_transfer(give_e_pair=nil, get_e_pair=nil, &block)
  if give_e_pair
    gc_orig = give_e_pair.charge
    give_e_pair.charge = gc_orig + 1
  end
  if get_e_pair
    rc_orig = get_e_pair.charge
    get_e_pair.charge = rc_orig - 1
  end

  reply = block.call(self)
  
  give_e_pair.charge = gc_orig if give_e_pair
  get_e_pair.charge = rc_orig if get_e_pair

  reply
end

#fragment(opts = {}) ⇒ Object

to ensure proper fragmentation, will add_h!(ph) first at the given ph an empty array is returned if there are no fragments generated.

:ph => 7.4
:uniq => false



150
151
152
153
154
155
156
157
158
159
160
161
162
163
164
165
166
167
168
169
170
171
172
173
174
175
176
177
178
179
180
181
182
183
184
185
186
187
188
189
190
191
192
193
194
195
196
197
198
199
200
201
202
203
204
205
206
207
208
209
210
211
212
213
214
215
216
217
218
219
220
221
222
223
224
225
226
227
228
229
230
231
232
233
234
235
236
237
238
239
240
241
242
243
244
245
246
247
248
249
250
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 150

def fragment(opts={})
  opts = DEFAULT_OPTIONS.merge(opts)
  opts[:rules].each do |rule| 
    raise ArgumentError, "bad rule: #{rule}" unless RULES.include?(rule)
  end

  had_hydrogens = self.h_added?

  self.correct_for_ph!(opts[:ph])
  self.remove_h!

  rules = opts[:rules]
  fragment_sets = []
  if rules.any? {|rule| CO_RULES.include?(rule) }
    putsv "matching C-O"
    self.each_match("CO").each do |_atoms|
      # note: this will *not* match C=O
      (carbon, oxygen) = _atoms
      carbon_nbrs = carbon.atoms.reject {|atom| atom == oxygen }
      c3_nbrs = carbon_nbrs.select {|atm| atm.type == 'C3' }
      # pulling this out here causes it to work incorrectly internally
      # (not sure why)
      #co_bond = carbon.get_bond(oxygen)

      case oxygen.bonds.size # non-hydrogen bonds
      when 1  # *must* be an alcohol or a carboxylic acid
        putsv "#{csmiles} oxygen has no other bonds besides C-O (alcohol or carboxylic acid)"
        if carbon.type == 'C3'
          if rules.include?(:sp3c_oxygen_double_bond_water_loss) 
            putsv "rule :sp3c_oxygen_double_bond_water_loss"
            fragment_sets.push *near_side_double_bond_break(carbon, oxygen)
          end
          if rules.include?(:alcohol_to_aldehyde)
              putsv "rule :alcohol_to_aldehyde"
            fragment_sets.push *alcohol_to_aldehyde(carbon, oxygen, carbon_nbrs)
          end
        elsif carbon.carboxyl_carbon?
          if rules.include?(:co2_loss)
            putsv "rule :co2_loss"
            if c3_nbr = c3_nbrs.first
              fragment_sets.push *co2_loss(carbon, oxygen, c3_nbr)
            end
          end
        end
      when 2
        putsv "#{csmiles} c-o & oxygen has 2 non-hydrogen bonds"
        oxygen_nbr = oxygen.atoms.reject {|atom| atom.idx == carbon.idx }.first
        if carbon.type == 'C3'
          if rules.include?(:peroxy_to_carboxy)
            fragment_sets.push *peroxy_to_carboxy(carbon, oxygen, carbon_nbrs, oxygen_nbr)
          end
          # ester and ethers (look *only* on close side for places to make
          # double bond)

          if oxygen_nbr.type == 'C3'
            putsv "oxygen nbr is C3"
            if rules.include?(:sp3c_oxygen_double_bond_far_side_sp3) 
              putsv "rule :sp3c_oxygen_double_bond_far_side_sp3"
              fragment_sets.push *near_side_double_bond_break(carbon, oxygen)
            end
            if rules.include?(:sp3c_oxygen_asymmetric_far_sp3)
              putsv "rule :sp3c_oxygen_asymmetric_far_sp3"
              # only returns a single frag set
              fragment_sets.push electrophile_snatches_electrons(carbon, oxygen)
            end
          end
          if oxygen_nbr.type == 'C2'
            if rules.include?(:sp3c_oxygen_double_bond_far_side_sp2)
              putsv "rule :sp3c_oxygen_double_bond_far_side_sp2"
              fragment_sets.push *near_side_double_bond_break(carbon, oxygen)
            end
          end
          # note: the case of a carboxy is found with CO search
        end
      end
    end
  end
  if rules.include?(:sp3c_nitrogen_double_bond)
    self.each_match("CN") do |_atoms|
      (carbon, nitrogen) = _atoms
      num_nitrogen_bonds = nitrogen.bonds.size
      case num_nitrogen_bonds
      when 2
        if carbon.type == 'C3'
          fragment_sets.push *near_side_double_bond_break(carbon, nitrogen)
        end
      end
    end
  end

  unless had_hydrogens
    fragment_sets.each {|set| set.each(&:remove_h!) }
    self.remove_h!
  end
  if opts[:uniq]
    # TODO: impelent properly
    #fragment_sets = fragment_sets.uniq_by(&:csmiles) 
    raise NotImplementedError
  end
  fragment_sets
end

#near_side_double_bond_break(carbon, electrophile) ⇒ Object 



91
92
93
94
95
96
97
98
99
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 91

def near_side_double_bond_break(carbon, electrophile)
  frag_sets = carbon.atoms.select {|atom| atom.type == "C3" }.map do |near_c3|
    frags = feint_double_bond(carbon.get_bond(near_c3)) do |_mol|
      frags = _mol.split(electrophile.get_bond(carbon))
      frags.map(&:add_h!)
    end
  end
  allowable_fragment_sets!(frag_sets)
end

#peroxy_to_carboxy(carbon, oxygen, carbon_nbrs, oxygen_nbr) ⇒ Object 



123
124
125
126
127
128
129
130
131
132
133
134
135
136
137
138
139
140
141
142
143
 
# File 'lib/rubabel/molecule/fragmentable.rb', line 123

def peroxy_to_carboxy(carbon, oxygen, carbon_nbrs, oxygen_nbr)
  if oxygen_nbr.el == :o # has a neighbor oxygen
    distal_o = oxygen_nbr
    if distal_o.bonds.size == 1  # this is a peroxy
      frag_sets = carbon_nbrs.select {|atom| atom.type == 'C3' }.map do |_atom|
        self.swap!(carbon, _atom, oxygen, distal_o)
        frags = feint_double_bond(carbon.get_bond(oxygen)) do |_mol|

          # we swapped the atoms so the bond to split off is now
          # attached to the oxygen
          frags = _mol.split(oxygen.get_bond(_atom))
          frags.map(&:add_h!)
        end
        self.swap!(carbon, distal_o, oxygen, _atom)
        frags
      end
      allowable_fragment_sets!(frag_sets)
    end
  end

end