Chem.MACCSkeys

Source Code for Module Chem.MACCSkeys

1 # $Id: MACCSkeys.py 775 2008-07-24 18:38:28Z glandrum $ 2 # 3 # Copyright (C) 2001-2008 greg Landrum and Rational Discovery LLC 4 # 5 # @@ All Rights Reserved @@ 6 # 7 """ SMARTS definitions for the publically available MACCS keys 8 and a MACCS fingerprinter 9 10 I compared the MACCS fingerprints generated here with those from two 11 other packages (not MDL, unfortunately). Of course there are 12 disagreements between the various fingerprints still, but I think 13 these definitions work pretty well. Some notes: 14 15 1) most of the differences have to do with aromaticity 16 2) there's a discrepancy sometimes because the current RDKit 17 definitions do not require multiple matches to be distinct. e.g. the 18 SMILES C(=O)CC(=O) can match the (hypothetical) key O=CC twice in my 19 definition. It's not clear to me what the correct behavior is. 20 3) Some keys are not fully defined in the MDL documentation 21 4) Two keys, 125 and 166, have to be done outside of SMARTS. 22 5) Key 1 (ISOTOPE) isn't defined 23 24 """ 25 import Chem 26 import DataStructs 27 # these are SMARTS patterns corresponding to the MDL MACCS keys 28 smartsPatts={ 29 1:('?',0), # ISOTOPE 30 #2:('[#103,#104,#105,#106,#107,#106,#109,#110,#111,#112]',0), # ISOTOPE Not complete 31 2:('[#103,#104]',0), # ISOTOPE Not complete 32 3:('[Ge,As,Se,Sn,Sb,Te,Tl,Pb,Bi]',0), # Group IVa,Va,VIa Periods 4-6 (Ge...) *NOTE* spec wrong 33 4:('[Ac,Th,Pa,U,Np,Pu,Am,Cm,Bk,Cf,Es,Fm,Md,No,Lr]',0), # actinide 34 5:('[Sc,Ti,Y,Zr,Hf]',0), # Group IIIB,IVB (Sc...) *NOTE* spec wrong 35 6:('[La,Ce,Pr,Nd,Pm,Sm,Eu,Gd,Tb,Dy,Ho,Er,Tm,Yb,Lu]',0), # Lanthanide 36 7:('[V,Cr,Mn,Nb,Mo,Tc,Ta,W,Re]',0), # Group VB,VIB,VIIB (V...) *NOTE* spec wrong 37 8:('[!#6;!#1]1~*~*~*~1',0), # QAAA@1 38 9:('[Fe,Co,Ni,Ru,Rh,Pd,Os,Ir,Pt]',0), # Group VIII (Fe...) 39 10:('[Be,Mg,Ca,Sr,Ba,Ra]',0), # Group IIa (Alkaline earth) 40 11:('*1~*~*~*~1',0), # 4M Ring 41 12:('[Cu,Zn,Ag,Cd,Au,Hg]',0), # Group IB,IIB (Cu..) 42 13:('[#8]~[#7](~[#6])~[#6]',0), # ON(C)C 43 14:('[#16]-[#16]',0), # S-S 44 15:('[#8]~[#6](~[#8])~[#8]',0), # OC(O)O 45 16:('[!#6;!#1]1~*~*~1',0), # QAA@1 46 17:('[#6]#[#6]',0), #CTC 47 18:('[B,Al,Ga,In,Tl]',0), # Group IIIA (B...) *NOTE* spec wrong 48 19:('*1~*~*~*~*~*~*~1',0), # 7M Ring 49 20:('[Si]',0), #Si 50 21:('[#6]=[#6](~[!#6;!#1])~[!#6;!#1]',0), # C=C(Q)Q 51 22:('*1~*~*~1',0), # 3M Ring 52 23:('[#7]~[#6](~[#8])~[#8]',0), # NC(O)O 53 24:('[#7]-[#8]',0), # N-O 54 25:('[#7]~[#6](~[#7])~[#7]',0), # NC(N)N 55 26:('[#6]=;@[#6](@*)@*',0), # C$=C($A)$A 56 27:('[I]',0), # I 57 28:('[!#6;!#1]~[CH2]~[!#6;!#1]',0), # QCH2Q 58 29:('[#15]',0),# P 59 30:('[#6]~[!#6;!#1](~[#6])(~[#6])~*',0), # CQ(C)(C)A 60 31:('[!#6;!#1]~[F,Cl,Br,I]',0), # QX 61 32:('[#6]~[#16]~[#7]',0), # CSN 62 33:('[#7]~[#16]',0), # NS 63 34:('[CH2]=*',0), # CH2=A 64 35:('[Li,Na,K,Rb,Cs,Fr]',0), # Group IA (Alkali Metal) 65 36:('[#16R]',0), # S Heterocycle 66 37:('[#7]~[#6](~[#8])~[#7]',0), # NC(O)N 67 38:('[#7]~[#6](~[#6])~[#7]',0), # NC(C)N 68 39:('[#8]~[#16](~[#8])~[#8]',0), # OS(O)O 69 40:('[#16]-[#8]',0), # S-O 70 41:('[#6]#[#7]',0), # CTN 71 42:('F',0), # F 72 43:('[!C;!c;!#1;!H0]~*~[!C;!c;!#1;!H0]',0), # QHAQH 73 44:('?',0), # OTHER 74 45:('[#6]=[#6]~[#7]',0), # C=CN 75 46:('Br',0), # BR 76 47:('[#16]~*~[#7]',0), # SAN 77 48:('[#8]~[!#6;!#1](~[#8])(~[#8])',0), # OQ(O)O 78 49:('[!+0]',0), # CHARGE 79 50:('[#6]=[#6](~[#6])~[#6]',0), # C=C(C)C 80 51:('[#6]~[#16]~[#8]',0), # CSO 81 52:('[#7]~[#7]',0), # NN 82 53:('[!#6;!#1;!H0]~*~*~*~[!#6;!#1;!H0]',0), # QHAAAQH 83 54:('[!#6;!#1;!H0]~*~*~[!#6;!#1;!H0]',0), # QHAAQH 84 55:('[#8]~[#16]~[#8]',0), #OSO 85 56:('[#8]~[#7](~[#8])~[#6]',0), # ON(O)C 86 57:('[#8R]',0), # O Heterocycle 87 58:('[!#6;!#1]~[#16]~[!#6;!#1]',0), # QSQ 88 59:('[#16]!:*:*',0), # Snot%A%A 89 60:('[#16]=[#8]',0), # S=O 90 61:('*~[#16](~*)~*',0), # AS(A)A 91 62:('*@*!@*@*',0), # A$!A$A 92 63:('[#7]=[#8]',0), # N=O 93 64:('*@*!@[#16]',0), # A$A!S 94 65:('c:n',0), # C%N 95 66:('[#6]~[#6](~[#6])(~[#6])~*',0), # CC(C)(C)A 96 67:('[!#6;!#1]~[#16]',0), # QS 97 68:('[!#6;!#1;!H0]~[!#6;!#1;!H0]',0), # QHQH (&...) FIX: incomplete definition 98 69:('[!#6;!#1]~[!#6;!#1;!H0]',0), # QQH 99 70:('[!#6;!#1]~[#7]~[!#6;!#1]',0), # QNQ 100 71:('[#7]~[#8]',0), # NO 101 72:('[#8]~*~*~[#8]',0), # OAAO 102 73:('[#16]=*',0), # S=A 103 74:('[CH3]~*~[CH3]',0), # CH3ACH3 104 75:('*!@[#7]@*',0), # A!N$A 105 76:('[#6]=[#6](~*)~*',0), # C=C(A)A 106 77:('[#7]~*~[#7]',0), # NAN 107 78:('[#6]=[#7]',0), # C=N 108 79:('[#7]~*~*~[#7]',0), # NAAN 109 80:('[#7]~*~*~*~[#7]',0), # NAAAN 110 81:('[#16]~*(~*)~*',0), # SA(A)A 111 82:('*~[CH2]~[!#6;!#1;!H0]',0), # ACH2QH 112 83:('[!#6;!#1]1~*~*~*~*~1',0), # QAAAA@1 113 84:('[NH2]',0), #NH2 114 85:('[#6]~[#7](~[#6])~[#6]',0), # CN(C)C 115 86:('[C;H2,H3][!#6;!#1][C;H2,H3]',0), # CH2QCH2 116 87:('[F,Cl,Br,I]!@*@*',0), # X!A$A 117 88:('[#16]',0), # S 118 89:('[#8]~*~*~*~[#8]',0), # OAAAO 119 90:('[$([!#6;!#1;!H0]~*~*~[CH2]~*),$([!#6;!#1;!H0;R]1@[R]@[R]@[CH2;R]1),$([!#6;!#1;!H0]~[R]1@[R]@[CH2;R]1)]',0), # QHAACH2A 120 91:('[$([!#6;!#1;!H0]~*~*~*~[CH2]~*),$([!#6;!#1;!H0;R]1@[R]@[R]@[R]@[CH2;R]1),$([!#6;!#1;!H0]~[R]1@[R]@[R]@[CH2;R]1),$([!#6;!#1;!H0]~*~[R]1@[R]@[CH2;R]1)]',0), # QHAAACH2A 121 92:('[#8]~[#6](~[#7])~[#6]',0), # OC(N)C 122 93:('[!#6;!#1]~[CH3]',0), # QCH3 123 94:('[!#6;!#1]~[#7]',0), # QN 124 95:('[#7]~*~*~[#8]',0), # NAAO 125 96:('*1~*~*~*~*~1',0), # 5 M ring 126 97:('[#7]~*~*~*~[#8]',0), # NAAAO 127 98:('[!#6;!#1]1~*~*~*~*~*~1',0), # QAAAAA@1 128 99:('[#6]=[#6]',0), # C=C 129 100:('*~[CH2]~[#7]',0), # ACH2N 130 101:('[$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1),$([R]@1@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]@[R]1)]',0), # 8M Ring or larger. This only handles up to ring sizes of 14 131 102:('[!#6;!#1]~[#8]',0), # QO 132 103:('Cl',0), # CL 133 104:('[!#6;!#1;!H0]~*~[CH2]~*',0), # QHACH2A 134 105:('*@*(@*)@*',0), # A$A($A)$A 135 106:('[!#6;!#1]~*(~[!#6;!#1])~[!#6;!#1]',0), # QA(Q)Q 136 107:('[F,Cl,Br,I]~*(~*)~*',0), # XA(A)A 137 108:('[CH3]~*~*~*~[CH2]~*',0), # CH3AAACH2A 138 109:('*~[CH2]~[#8]',0), # ACH2O 139 110:('[#7]~[#6]~[#8]',0), # NCO 140 111:('[#7]~*~[CH2]~*',0), # NACH2A 141 112:('*~*(~*)(~*)~*',0), # AA(A)(A)A 142 113:('[#8]!:*:*',0), # Onot%A%A 143 114:('[CH3]~[CH2]~*',0), # CH3CH2A 144 115:('[CH3]~*~[CH2]~*',0), # CH3ACH2A 145 116:('[$([CH3]~*~*~[CH2]~*),$([CH3]~*1~*~[CH2]1)]',0), # CH3AACH2A 146 117:('[#7]~*~[#8]',0), # NAO 147 118:('[$(*~[CH2]~[CH2]~*),$(*1~[CH2]~[CH2]1)]',1), # ACH2CH2A > 1 148 119:('[#7]=*',0), # N=A 149 120:('[!#6;R]',1), # Heterocyclic atom > 1 (&...) FIX: incomplete definition 150 121:('[#7;R]',0), # N Heterocycle 151 122:('*~[#7](~*)~*',0), # AN(A)A 152 123:('[#8]~[#6]~[#8]',0), # OCO 153 124:('[!#6;!#1]~[!#6;!#1]',0), # QQ 154 125:('?',0), # Aromatic Ring > 1 155 126:('*!@[#8]!@*',0), # A!O!A 156 127:('*@*!@[#8]',1), # A$A!O > 1 (&...) FIX: incomplete definition 157 128:('[$(*~[CH2]~*~*~*~[CH2]~*),$([R]1@[CH2;R]@[R]@[R]@[R]@[CH2;R]1),$(*~[CH2]~[R]1@[R]@[R]@[CH2;R]1),$(*~[CH2]~*~[R]1@[R]@[CH2;R]1)]',0), # ACH2AAACH2A 158 129:('[$(*~[CH2]~*~*~[CH2]~*),$([R]1@[CH2]@[R]@[R]@[CH2;R]1),$(*~[CH2]~[R]1@[R]@[CH2;R]1)]',0), # ACH2AACH2A 159 130:('[!#6;!#1]~[!#6;!#1]',1), # QQ > 1 (&...) FIX: incomplete definition 160 131:('[!#6;!#1;!H0]',1), # QH > 1 161 132:('[#8]~*~[CH2]~*',0), # OACH2A 162 133:('*@*!@[#7]',0), # A$A!N 163 134:('[F,Cl,Br,I]',0), # X (HALOGEN) 164 135:('[#7]!:*:*',0), # Nnot%A%A 165 136:('[#8]=*',1), # O=A>1 166 137:('[!C;!c;R]',0), # Heterocycle 167 138:('[!#6;!#1]~[CH2]~*',1), # QCH2A>1 (&...) FIX: incomplete definition 168 139:('[O;!H0]',0), # OH 169 140:('[#8]',3), # O > 3 (&...) FIX: incomplete definition 170 141:('[CH3]',2), # CH3 > 2 (&...) FIX: incomplete definition 171 142:('[#7]',1), # N > 1 172 143:('*@*!@[#8]',0), # A$A!O 173 144:('*!:*:*!:*',0), # Anot%A%Anot%A 174 145:('*1~*~*~*~*~*~1',1), # 6M ring > 1 175 146:('[#8]',2), # O > 2 176 147:('[$(*~[CH2]~[CH2]~*),$([R]1@[CH2;R]@[CH2;R]1)]',0), # ACH2CH2A 177 148:('*~[!#6;!#1](~*)~*',0), # AQ(A)A 178 149:('[C;H3,H4]',1), # CH3 > 1 179 150:('*!@*@*!@*',0), # A!A$A!A 180 151:('[#7;!H0]',0), # NH 181 152:('[#8]~[#6](~[#6])~[#6]',0), # OC(C)C 182 153:('[!#6;!#1]~[CH2]~*',0), # QCH2A 183 154:('[#6]=[#8]',0), # C=O 184 155:('*!@[CH2]!@*',0), # A!CH2!A 185 156:('[#7]~*(~*)~*',0), # NA(A)A 186 157:('[#6]-[#8]',0), # C-O 187 158:('[#6]-[#7]',0), # C-N 188 159:('[#8]',1), # O>1 189 160:('[C;H3,H4]',0), #CH3 190 161:('[#7]',0), # N 191 162:('a',0), # Aromatic 192 163:('*1~*~*~*~*~*~1',0), # 6M Ring 193 164:('[#8]',0), # O 194 165:('[R]',0), # Ring 195 166:('?',0), # Fragments FIX: this can't be done in SMARTS 196 } 197 198 maccsKeys = None 199

200 -def _InitKeys(keyList,keyDict):

201 """ *Internal Use Only* 202 203 generates SMARTS patterns for the keys, run once 204 205 """ 206 assert len(keyList) == len(keyDict.keys()),'length mismatch' 207 for key in keyDict.keys(): 208 patt,count = keyDict[key] 209 if patt != '?': 210 try: 211 sma = Chem.MolFromSmarts(patt) 212 except: 213 sma = None 214 if not sma: 215 print 'SMARTS parser error for key #%d: %s'%(key,patt) 216 else: 217 keyList[key-1] = sma,count

218

219 -def GenMACCSKeys(mol,**kwargs):

220 """ generates the MACCS fingerprint for a molecules 221 222 **Arguments** 223 224 - mol: the molecule to be fingerprinted 225 226 - any extra keyword arguments are ignored 227 228 **Returns** 229 230 a _DataStructs.SparseBitVect_ containing the fingerprint. 231 232 >>> m = Chem.MolFromSmiles('CNO') 233 >>> bv = GenMACCSKeys(m) 234 >>> tuple(bv.GetOnBits()) 235 (24, 68, 69, 71, 93, 94, 102, 124, 131, 139, 151, 158, 160, 161, 164) 236 >>> bv = GenMACCSKeys(Chem.MolFromSmiles('CCC')) 237 >>> tuple(bv.GetOnBits()) 238 (74, 114, 149, 155, 160) 239 240 """ 241 global maccsKeys 242 if maccsKeys is None: 243 maccsKeys = [(None,0)]*len(smartsPatts.keys()) 244 _InitKeys(maccsKeys,smartsPatts) 245 ctor=kwargs.get('ctor',DataStructs.SparseBitVect) 246 247 res = ctor(len(maccsKeys)+1) 248 for i,(patt,count) in enumerate(maccsKeys): 249 if patt is not None: 250 if count==0: 251 res[i+1] = mol.HasSubstructMatch(patt) 252 else: 253 matches = mol.GetSubstructMatches(patt) 254 if len(matches) > count: 255 res[i+1] = 1 256 elif (i+1)==125: 257 # special case: num aromatic rings > 1 258 ri = mol.GetRingInfo() 259 nArom=0 260 res[125]=0 261 for ring in ri.BondRings(): 262 isArom=True 263 for bondIdx in ring: 264 if not mol.GetBondWithIdx(bondIdx).GetIsAromatic(): 265 isArom=False 266 break 267 if isArom: 268 nArom+=1 269 if nArom>1: 270 res[125]=1 271 break 272 elif (i+1)==166: 273 res[166]=0 274 # special case: num frags > 1 275 if len(Chem.GetMolFrags(mol))>1: 276 res[166]=1 277 278 return res

279 280 FingerprintMol = GenMACCSKeys 281 282 #------------------------------------ 283 # 284 # doctest boilerplate 285 #

286 -def _test():

287 import doctest,sys 288 return doctest.testmod(sys.modules["__main__"])

289 290 if __name__ == '__main__': 291 import sys 292 failed,tried = _test() 293 sys.exit(failed) 294