rdkit.Chem.MACCSkeys

Source Code for Module rdkit.Chem.MACCSkeys

1 # $Id: MACCSkeys.py 1733 2011-05-27 03:19:13Z glandrum $ 2 # 3 # Copyright (C) 2001-2011 greg Landrum and Rational Discovery LLC 4 # 5 # @@ All Rights Reserved @@ 6 # This file is part of the RDKit. 7 # The contents are covered by the terms of the BSD license 8 # which is included in the file license.txt, found at the root 9 # of the RDKit source tree. 10 # 11 """ SMARTS definitions for the publically available MACCS keys 12 and a MACCS fingerprinter 13 14 I compared the MACCS fingerprints generated here with those from two 15 other packages (not MDL, unfortunately). Of course there are 16 disagreements between the various fingerprints still, but I think 17 these definitions work pretty well. Some notes: 18 19 1) most of the differences have to do with aromaticity 20 2) there's a discrepancy sometimes because the current RDKit 21 definitions do not require multiple matches to be distinct. e.g. the 22 SMILES C(=O)CC(=O) can match the (hypothetical) key O=CC twice in my 23 definition. It's not clear to me what the correct behavior is. 24 3) Some keys are not fully defined in the MDL documentation 25 4) Two keys, 125 and 166, have to be done outside of SMARTS. 26 5) Key 1 (ISOTOPE) isn't defined 27 28 Rev history: 29 2006 (gl): Original open-source release 30 May 2011 (gl): Update some definitions based on feedback from Andrew Dalke 31 32 """ 33 from rdkit import Chem 34 from rdkit import DataStructs 35 # these are SMARTS patterns corresponding to the MDL MACCS keys 36 smartsPatts={ 37 1:('?',0), # ISOTOPE 38 #2:('[#104,#105,#106,#107,#106,#109,#110,#111,#112]',0), # atomic num >103 Not complete 39 2:('[#104]',0), # limit the above def'n since the RDKit only accepts up to #104 40 3:('[#32,#33,#34,#50,#51,#52,#82,#83,#84]',0), # Group IVa,Va,VIa Rows 4-6 41 4:('[Ac,Th,Pa,U,Np,Pu,Am,Cm,Bk,Cf,Es,Fm,Md,No,Lr]',0), # actinide 42 5:('[Sc,Ti,Y,Zr,Hf]',0), # Group IIIB,IVB (Sc...) 43 6:('[La,Ce,Pr,Nd,Pm,Sm,Eu,Gd,Tb,Dy,Ho,Er,Tm,Yb,Lu]',0), # Lanthanide 44 7:('[V,Cr,Mn,Nb,Mo,Tc,Ta,W,Re]',0), # Group VB,VIB,VIIB 45 8:('[!#6;!#1]1~*~*~*~1',0), # QAAA@1 46 9:('[Fe,Co,Ni,Ru,Rh,Pd,Os,Ir,Pt]',0), # Group VIII (Fe...) 47 10:('[Be,Mg,Ca,Sr,Ba,Ra]',0), # Group IIa (Alkaline earth) 48 11:('*1~*~*~*~1',0), # 4M Ring 49 12:('[Cu,Zn,Ag,Cd,Au,Hg]',0), # Group IB,IIB (Cu..) 50 13:('[#8]~[#7](~[#6])~[#6]',0), # ON(C)C 51 14:('[#16]-[#16]',0), # S-S 52 15:('[#8]~[#6](~[#8])~[#8]',0), # OC(O)O 53 16:('[!#6;!#1]1~*~*~1',0), # QAA@1 54 17:('[#6]#[#6]',0), #CTC 55 18:('[#5,#13,#31,#49,#81]',0), # Group IIIA (B...) 56 19:('*1~*~*~*~*~*~*~1',0), # 7M Ring 57 20:('[Si]',0), #Si 58 21:('[#6]=[#6](~[!#6;!#1])~[!#6;!#1]',0), # C=C(Q)Q 59 22:('*1~*~*~1',0), # 3M Ring 60 23:('[#7]~[#6](~[#8])~[#8]',0), # NC(O)O 61 24:('[#7]-[#8]',0), # N-O 62 25:('[#7]~[#6](~[#7])~[#7]',0), # NC(N)N 63 26:('[#6]=;@[#6](@*)@*',0), # C$=C($A)$A 64 27:('[I]',0), # I 65 28:('[!#6;!#1]~[CH2]~[!#6;!#1]',0), # QCH2Q 66 29:('[#15]',0),# P 67 30:('[#6]~[!#6;!#1](~[#6])(~[#6])~*',0), # CQ(C)(C)A 68 31:('[!#6;!#1]~[F,Cl,Br,I]',0), # QX 69 32:('[#6]~[#16]~[#7]',0), # CSN 70 33:('[#7]~[#16]',0), # NS 71 34:('[CH2]=*',0), # CH2=A 72 35:('[Li,Na,K,Rb,Cs,Fr]',0), # Group IA (Alkali Metal) 73 36:('[#16R]',0), # S Heterocycle 74 37:('[#7]~[#6](~[#8])~[#7]',0), # NC(O)N 75 38:('[#7]~[#6](~[#6])~[#7]',0), # NC(C)N 76 39:('[#8]~[#16](~[#8])~[#8]',0), # OS(O)O 77 40:('[#16]-[#8]',0), # S-O 78 41:('[#6]#[#7]',0), # CTN 79 42:('F',0), # F 80 43:('[!#6;!#1;!H0]~*~[!#6;!#1;!H0]',0), # QHAQH 81 44:('?',0), # OTHER 82 45:('[#6]=[#6]~[#7]',0), # C=CN 83 46:('Br',0), # BR 84 47:('[#16]~*~[#7]',0), # SAN 85 48:('[#8]~[!#6;!#1](~[#8])(~[#8])',0), # OQ(O)O 86 49:('[!+0]',0), # CHARGE 87 50:('[#6]=[#6](~[#6])~[#6]',0), # C=C(C)C 88 51:('[#6]~[#16]~[#8]',0), # CSO 89 52:('[#7]~[#7]',0), # NN 90 53:('[!#6;!#1;!H0]~*~*~*~[!#6;!#1;!H0]',0), # QHAAAQH 91 54:('[!#6;!#1;!H0]~*~*~[!#6;!#1;!H0]',0), # QHAAQH 92 55:('[#8]~[#16]~[#8]',0), #OSO 93 56:('[#8]~[#7](~[#8])~[#6]',0), # ON(O)C 94 57:('[#8R]',0), # O Heterocycle 95 58:('[!#6;!#1]~[#16]~[!#6;!#1]',0), # QSQ 96 59:('[#16]!:*:*',0), # Snot%A%A 97 60:('[#16]=[#8]',0), # S=O 98 61:('*~[#16](~*)~*',0), # AS(A)A 99 62:('*@*!@*@*',0), # A$!A$A 100 63:('[#7]=[#8]',0), # N=O 101 64:('*@*!@[#16]',0), # A$A!S 102 65:('c:n',0), # C%N 103 66:('[#6]~[#6](~[#6])(~[#6])~*',0), # CC(C)(C)A 104 67:('[!#6;!#1]~[#16]',0), # QS 105 68:('[!#6;!#1;!H0]~[!#6;!#1;!H0]',0), # QHQH (&...) SPEC Incomplete 106 69:('[!#6;!#1]~[!#6;!#1;!H0]',0), # QQH 107 70:('[!#6;!#1]~[#7]~[!#6;!#1]',0), # QNQ 108 71:('[#7]~[#8]',0), # NO 109 72:('[#8]~*~*~[#8]',0), # OAAO 110 73:('[#16]=*',0), # S=A 111 74:('[CH3]~*~[CH3]',0), # CH3ACH3 112 75:('*!@[#7]@*',0), # A!N$A 113 76:('[#6]=[#6](~*)~*',0), # C=C(A)A 114 77:('[#7]~*~[#7]',0), # NAN 115 78:('[#6]=[#7]',0), # C=N 116 79:('[#7]~*~*~[#7]',0), # NAAN 117 80:('[#7]~*~*~*~[#7]',0), # NAAAN 118 81:('[#16]~*(~*)~*',0), # SA(A)A 119 82:('*~[CH2]~[!#6;!#1;!H0]',0), # ACH2QH 120 83:('[!#6;!#1]1~*~*~*~*~1',0), # QAAAA@1 121 84:('[NH2]',0), #NH2 122 85:('[#6]~[#7](~[#6])~[#6]',0), # CN(C)C 123 86:('[C;H2,H3][!#6;!#1][C;H2,H3]',0), # CH2QCH2 124 87:('[F,Cl,Br,I]!@*@*',0), # X!A$A 125 88:('[#16]',0), # S 126 89:('[#8]~*~*~*~[#8]',0), # OAAAO 127 90:('[$([!#6;!#1;!H0]~*~*~[CH2]~*),$([!#6;!#1;!H0;R]1@[R]@[R]@[CH2;R]1),$([!#6;!#1;!H0]~[R]1@[R]@[CH2;R]1)]',0), # QHAACH2A 128 91:('[$([!#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),$([!#6;!#1;!H0]~*~[R]1@[R]@[CH2;R]1)]',0), # QHAAACH2A 129 92:('[#8]~[#6](~[#7])~[#6]',0), # OC(N)C 130 93:('[!#6;!#1]~[CH3]',0), # QCH3 131 94:('[!#6;!#1]~[#7]',0), # QN 132 95:('[#7]~*~*~[#8]',0), # NAAO 133 96:('*1~*~*~*~*~1',0), # 5 M ring 134 97:('[#7]~*~*~*~[#8]',0), # NAAAO 135 98:('[!#6;!#1]1~*~*~*~*~*~1',0), # QAAAAA@1 136 99:('[#6]=[#6]',0), # C=C 137 100:('*~[CH2]~[#7]',0), # ACH2N 138 101:('[$([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 14 139 102:('[!#6;!#1]~[#8]',0), # QO 140 103:('Cl',0), # CL 141 104:('[!#6;!#1;!H0]~*~[CH2]~*',0), # QHACH2A 142 105:('*@*(@*)@*',0), # A$A($A)$A 143 106:('[!#6;!#1]~*(~[!#6;!#1])~[!#6;!#1]',0), # QA(Q)Q 144 107:('[F,Cl,Br,I]~*(~*)~*',0), # XA(A)A 145 108:('[CH3]~*~*~*~[CH2]~*',0), # CH3AAACH2A 146 109:('*~[CH2]~[#8]',0), # ACH2O 147 110:('[#7]~[#6]~[#8]',0), # NCO 148 111:('[#7]~*~[CH2]~*',0), # NACH2A 149 112:('*~*(~*)(~*)~*',0), # AA(A)(A)A 150 113:('[#8]!:*:*',0), # Onot%A%A 151 114:('[CH3]~[CH2]~*',0), # CH3CH2A 152 115:('[CH3]~*~[CH2]~*',0), # CH3ACH2A 153 116:('[$([CH3]~*~*~[CH2]~*),$([CH3]~*1~*~[CH2]1)]',0), # CH3AACH2A 154 117:('[#7]~*~[#8]',0), # NAO 155 118:('[$(*~[CH2]~[CH2]~*),$(*1~[CH2]~[CH2]1)]',1), # ACH2CH2A > 1 156 119:('[#7]=*',0), # N=A 157 120:('[!#6;R]',1), # Heterocyclic atom > 1 (&...) Spec Incomplete 158 121:('[#7;R]',0), # N Heterocycle 159 122:('*~[#7](~*)~*',0), # AN(A)A 160 123:('[#8]~[#6]~[#8]',0), # OCO 161 124:('[!#6;!#1]~[!#6;!#1]',0), # QQ 162 125:('?',0), # Aromatic Ring > 1 163 126:('*!@[#8]!@*',0), # A!O!A 164 127:('*@*!@[#8]',1), # A$A!O > 1 (&...) Spec Incomplete 165 128:('[$(*~[CH2]~*~*~*~[CH2]~*),$([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 166 129:('[$(*~[CH2]~*~*~[CH2]~*),$([R]1@[CH2]@[R]@[R]@[CH2;R]1),$(*~[CH2]~[R]1@[R]@[CH2;R]1)]',0), # ACH2AACH2A 167 130:('[!#6;!#1]~[!#6;!#1]',1), # QQ > 1 (&...) Spec Incomplete 168 131:('[!#6;!#1;!H0]',1), # QH > 1 169 132:('[#8]~*~[CH2]~*',0), # OACH2A 170 133:('*@*!@[#7]',0), # A$A!N 171 134:('[F,Cl,Br,I]',0), # X (HALOGEN) 172 135:('[#7]!:*:*',0), # Nnot%A%A 173 136:('[#8]=*',1), # O=A>1 174 137:('[!C;!c;R]',0), # Heterocycle 175 138:('[!#6;!#1]~[CH2]~*',1), # QCH2A>1 (&...) Spec Incomplete 176 139:('[O;!H0]',0), # OH 177 140:('[#8]',3), # O > 3 (&...) Spec Incomplete 178 141:('[CH3]',2), # CH3 > 2 (&...) Spec Incomplete 179 142:('[#7]',1), # N > 1 180 143:('*@*!@[#8]',0), # A$A!O 181 144:('*!:*:*!:*',0), # Anot%A%Anot%A 182 145:('*1~*~*~*~*~*~1',1), # 6M ring > 1 183 146:('[#8]',2), # O > 2 184 147:('[$(*~[CH2]~[CH2]~*),$([R]1@[CH2;R]@[CH2;R]1)]',0), # ACH2CH2A 185 148:('*~[!#6;!#1](~*)~*',0), # AQ(A)A 186 149:('[C;H3,H4]',1), # CH3 > 1 187 150:('*!@*@*!@*',0), # A!A$A!A 188 151:('[#7;!H0]',0), # NH 189 152:('[#8]~[#6](~[#6])~[#6]',0), # OC(C)C 190 153:('[!#6;!#1]~[CH2]~*',0), # QCH2A 191 154:('[#6]=[#8]',0), # C=O 192 155:('*!@[CH2]!@*',0), # A!CH2!A 193 156:('[#7]~*(~*)~*',0), # NA(A)A 194 157:('[#6]-[#8]',0), # C-O 195 158:('[#6]-[#7]',0), # C-N 196 159:('[#8]',1), # O>1 197 160:('[C;H3,H4]',0), #CH3 198 161:('[#7]',0), # N 199 162:('a',0), # Aromatic 200 163:('*1~*~*~*~*~*~1',0), # 6M Ring 201 164:('[#8]',0), # O 202 165:('[R]',0), # Ring 203 166:('?',0), # Fragments FIX: this can't be done in SMARTS 204 } 205 206 maccsKeys = None 207

208 -def _InitKeys(keyList,keyDict):

209 """ *Internal Use Only* 210 211 generates SMARTS patterns for the keys, run once 212 213 """ 214 assert len(keyList) == len(keyDict.keys()),'length mismatch' 215 for key in keyDict.keys(): 216 patt,count = keyDict[key] 217 if patt != '?': 218 try: 219 sma = Chem.MolFromSmarts(patt) 220 except: 221 sma = None 222 if not sma: 223 print 'SMARTS parser error for key #%d: %s'%(key,patt) 224 else: 225 keyList[key-1] = sma,count

226

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

228 """ generates the MACCS fingerprint for a molecules 229 230 **Arguments** 231 232 - mol: the molecule to be fingerprinted 233 234 - any extra keyword arguments are ignored 235 236 **Returns** 237 238 a _DataStructs.SparseBitVect_ containing the fingerprint. 239 240 >>> m = Chem.MolFromSmiles('CNO') 241 >>> bv = GenMACCSKeys(m) 242 >>> tuple(bv.GetOnBits()) 243 (24, 68, 69, 71, 93, 94, 102, 124, 131, 139, 151, 158, 160, 161, 164) 244 >>> bv = GenMACCSKeys(Chem.MolFromSmiles('CCC')) 245 >>> tuple(bv.GetOnBits()) 246 (74, 114, 149, 155, 160) 247 248 """ 249 global maccsKeys 250 if maccsKeys is None: 251 maccsKeys = [(None,0)]*len(smartsPatts.keys()) 252 _InitKeys(maccsKeys,smartsPatts) 253 ctor=kwargs.get('ctor',DataStructs.SparseBitVect) 254 255 res = ctor(len(maccsKeys)+1) 256 for i,(patt,count) in enumerate(maccsKeys): 257 if patt is not None: 258 if count==0: 259 res[i+1] = mol.HasSubstructMatch(patt) 260 else: 261 matches = mol.GetSubstructMatches(patt) 262 if len(matches) > count: 263 res[i+1] = 1 264 elif (i+1)==125: 265 # special case: num aromatic rings > 1 266 ri = mol.GetRingInfo() 267 nArom=0 268 res[125]=0 269 for ring in ri.BondRings(): 270 isArom=True 271 for bondIdx in ring: 272 if not mol.GetBondWithIdx(bondIdx).GetIsAromatic(): 273 isArom=False 274 break 275 if isArom: 276 nArom+=1 277 if nArom>1: 278 res[125]=1 279 break 280 elif (i+1)==166: 281 res[166]=0 282 # special case: num frags > 1 283 if len(Chem.GetMolFrags(mol))>1: 284 res[166]=1 285 286 return res

287 288 FingerprintMol = GenMACCSKeys 289 290 #------------------------------------ 291 # 292 # doctest boilerplate 293 #

294 -def _test():

295 import doctest,sys 296 return doctest.testmod(sys.modules["__main__"])

297 298 if __name__ == '__main__': 299 import sys 300 failed,tried = _test() 301 sys.exit(failed) 302