Add description of functional groups to fingerprintformat and tidy up fingerprints FP3 and FP4

2025-02-25 18:55:23 -06:00 · 2007-08-30 13:41:36 +00:00
parent b072abea4f
commit ed0bee278b
8 changed files with 192 additions and 147 deletions
--- a/3
+++ b/3
@@ -1,3 +1,6 @@
+2007-08-30  Chris Morley
+
+	
 2007-08-27  Jean Brefort  <jean@acer>

 	* include/openbabel/generic.h: make OBUnitCell use SpaceGroup.
--- a/data/SMARTS_InteLigand.txt
+++ b/data/SMARTS_InteLigand.txt
@@ -2,9 +2,11 @@
 #              SMARTS Patterns for Functional Group Classification 
 #
 #              written by Christian Laggner 
-#              Copyright Inte:Ligand Software-Entwicklungs und Consulting GmbH
-#              Released under the Lesser General Public License (LGPL license) 
-#                      http://www.gnu.org/copyleft/lesser.html 
+#              Copyright 2005 Inte:Ligand Software-Entwicklungs und Consulting GmbH
+#
+#              Released under the Lesser General Public License (LGPL license)
+#              see http://www.gnu.org/copyleft/lesser.html
+#              Modified from Version 221105
 #####################################################################################################

 # General Stuff:
@@ -31,7 +33,7 @@ Secondary_carbon: [CX4H2]([#6])[#6]

 Tertiary_carbon: [CX4H1]([#6])([#6])[#6]

-Quartary_carbon: [CX4]([#6])([#6])([#6])[#6]
+Quaternary_carbon: [CX4]([#6])([#6])([#6])[#6]


 # I.1.2 C-C double and Triple Bonds
@@ -85,7 +87,7 @@ Alkylarylether: [OX2](c)[CX4;!$(C([OX2])[O,S,#7,#15,F,Cl,Br,I])]

 Diarylether: [c][OX2][c]

-Alkylarylthioether: [SX2](c)[CX4;!$(C([OX2])[O,S,#7,#15,F,Cl,Br,I])
+Alkylarylthioether: [SX2](c)[CX4;!$(C([OX2])[O,S,#7,#15,F,Cl,Br,I])]

 Diarylthioether: [c][SX2][c]

@@ -94,36 +96,36 @@ Oxonium: [O+;!$([O]~[!#6]);!$([S]*~[#7,#8,#15,#16])]

 # I.2.3 Amines

-Amine: [NX3,NX4+;!$([N]~[!#6]);!$([N]*~[#7,#8,#15,#16])]
+Amine: [NX3+0,NX4+;!$([N]~[!#6]);!$([N]*~[#7,#8,#15,#16])]
 # hits all amines (prim/sec/tert/quart), including ammonium salts, also enamines, but not amides, imides, aminals, ...

 # the following amines include also the protonated forms 

-Primary_aliph_amine: [NX3H2,NX4H3+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]
+Primary_aliph_amine: [NX3H2+0,NX4H3+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]

-Secondary_aliph_amine: [NX3H1,NX4H2+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]
+Secondary_aliph_amine: [NX3H1+0,NX4H2+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]

-Tertiary_aliph_amine: [NX3H0,NX4H1+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]
+Tertiary_aliph_amine: [NX3H0+0,NX4H1+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]

-Quartary_aliph_ammonium: [NX4H0+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]
+Quaternary_aliph_ammonium: [NX4H0+;!$([N][!C]);!$([N]*~[#7,#8,#15,#16])]

-Primary_arom_amine: [NX3H2,NX4H3+]c
+Primary_arom_amine: [NX3H2+0,NX4H3+]c

-Secondary_arom_amine: [NX3H1,NX4H2+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]
+Secondary_arom_amine: [NX3H1+0,NX4H2+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]

-Tertiary_arom_amine: [NX3H0,NX4H1+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]
+Tertiary_arom_amine: [NX3H0+0,NX4H1+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]

-Quartary_arom_ammonium: [NX4H0+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]
+Quaternary_arom_ammonium: [NX4H0+;!$([N][!c]);!$([N]*~[#7,#8,#15,#16])]

-Secondary_mixed_amine: [NX3H1,NX4H2+;$([N]([c])[C]);!$([N]*~[#7,#8,#15,#16])]
+Secondary_mixed_amine: [NX3H1+0,NX4H2+;$([N]([c])[C]);!$([N]*~[#7,#8,#15,#16])]

-Tertiary_mixed_amine: [NX3H0,NX4H1+;$([N]([c])([C])[#6]);!$([N]*~[#7,#8,#15,#16])]
+Tertiary_mixed_amine: [NX3H0+0,NX4H1+;$([N]([c])([C])[#6]);!$([N]*~[#7,#8,#15,#16])]

-Quartary_mixed_ammonium: [NX4H0+;$([N]([c])([C])[#6][#6]);!$([N]*~[#7,#8,#15,#16])]
+Quaternary_mixed_ammonium: [NX4H0+;$([N]([c])([C])[#6][#6]);!$([N]*~[#7,#8,#15,#16])]

-Ammonium: [N+;!$([N]~[!#6]);!$([N]*~[#7,#8,#15,#16])]
-# only C and H substituents allowed. NX4+ or Nv4+ is not recognized by Daylight's
-# depictmatch if less than four C are present...
+Ammonium: [N+;!$([N]~[!#6]);!$(N=*);!$([N]*~[#7,#8,#15,#16])]
+# only C and H substituents allowed. Quaternary or protonated amines
+# NX4+ or Nv4+ is not recognized by Daylight's depictmatch if less than four C are present


 # I.2.4 Others
@@ -177,7 +179,7 @@ Thioketone: [#6][CX3](=[SX1])[#6]
 Imine: [NX2;$([N][#6]),$([NH]);!$([N][CX3]=[#7,#8,#15,#16])]=[CX3;$([CH2]),$([CH][#6]),$([C]([#6])[#6])]
 # nitrogen is not part of an amidelike strukture, nor of an aromatic ring, but can be part of an aminal or similar

-Immonium: [N+;!$([N][!#6]);!$([N][CX3]=[#7,#8,#15,#16])]]
+Immonium: [NX3+;!$([N][!#6]);!$([N][CX3]=[#7,#8,#15,#16])]

 Oxime: [NX2](=[CX3;$([CH2]),$([CH][#6]),$([C]([#6])[#6])])[OX2H]

@@ -337,13 +339,13 @@ Thiolactam: [#6R][#6X3R](=[SX1])[#7X3;$([H1][#6;!$(C=[O,N,S])]),$([H0]([#6;!$(C=
 Oximester: [#6X3;$([H0][#6]),$([H1])](=[OX1])[#8X2][#7X2]=,:[#6X3;$([H0]([#6])[#6]),$([H1][#6]),$([H2])]
 # may also be part of a ring / aromatic

-Amidine: [NX3;!$(NC=[O,S])][CX3;$([CH],$([C][#6])]=[NX2;!$(NC=[O,S])]
+Amidine: [NX3;!$(NC=[O,S])][CX3;$([CH]),$([C][#6])]=[NX2;!$(NC=[O,S])]
 # only basic amidines, not as part of aromatic ring (e.g. imidazole)

 Hydroxamic_acid: [CX3;$([H0][#6]),$([H1])](=[OX1])[#7X3;$([H1]),$([H0][#6;!$(C=[O,N,S])])][$([OX2H]),$([OX1-])]

-Hydroxamic_acid_ester: [CX3;$([H0][#6]),$([H1])](=[OX1])[#7X3;$([H1]),$([H0][#6;!$(C=[O,N,S])])][OX2][#6;!$(C=[O,N,S])])]
-# #does not hit anhydrides of carboxylic acids withs hydroxamic acids
+Hydroxamic_acid_ester: [CX3;$([H0][#6]),$([H1])](=[OX1])[#7X3;$([H1]),$([H0][#6;!$(C=[O,N,S])])][OX2][#6;!$(C=[O,N,S])]
+# does not hit anhydrides of carboxylic acids withs hydroxamic acids


 Imidoacid: [CX3R0;$([H0][#6]),$([H1])](=[NX2;$([H1]),$([H0][#6;!$(C=[O,N,S])])])[$([OX2H]),$([OX1-])]
@@ -483,15 +485,15 @@ Semicarbazide: [#7X3][#7X3][#6X3]([#7X3;!$([#7][#7])])=[OX1]

 Carbazide: [#7X3][#7X3][#6X3]([#7X3][#7X3])=[OX1]

-Semicarbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3;!$([#7][#7]))=[OX1]
+Semicarbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3;!$([#7][#7])])=[OX1]

 Carbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3][#7X3])=[OX1]

-Thiosemicarbazide: [#7X3][#7X3][#6X3]([#7X3;!$([#7][#7])=[SX1]
+Thiosemicarbazide: [#7X3][#7X3][#6X3]([#7X3;!$([#7][#7])])=[SX1]

 Thiocarbazide: [#7X3][#7X3][#6X3]([#7X3][#7X3])=[SX1]

-Thiosemicarbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3;!$([#7][#7])=[SX1]
+Thiosemicarbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3;!$([#7][#7])])=[SX1]

 Thiocarbazone: [#7X2](=[#6])[#7X3][#6X3]([#7X3][#7X3])=[SX1]

@@ -588,7 +590,7 @@ Diazonium: [#6][NX2+]#[NX1]

 Nitrosamine: [#7;!$(N*=O)][NX2]=[OX1]

-Nitrosamide: [NX2](=[OX1])[N-*=O]
+Nitrosamide: [NX2](=[OX1])N-*=O
 # includes nitrososulfonamides

 N-Oxide: [$([#7+][OX1-]),$([#7v5]=[OX1]);!$([#7](~[O])~[O]);!$([#7]=[#7])] 
@@ -680,7 +682,7 @@ Sulfenic_derivative: [SX2;$([H1]),$([H0][#6])][!#6]
 Phosphine: [PX3;$([H3]),$([H2][#6]),$([H1]([#6])[#6]),$([H0]([#6])([#6])[#6])]
 # similar to amine, but less restrictive: includes also amide- and aminal-analogues 

-Phosphine_oxide: [PX4;$([H3]=[OX]),$([H2](=[OX])[#6]),$([H1](=[OX])([#6])[#6]),$([H0](=[OX])([#6])([#6])[#6])]
+Phosphine_oxide: [PX4;$([H3]=[OX1]),$([H2](=[OX1])[#6]),$([H1](=[OX1])([#6])[#6]),$([H0](=[OX1])([#6])([#6])[#6])]

 Phosphonium: [P+;!$([P]~[!#6]);!$([P]*~[#7,#8,#15,#16])]
 # similar to Ammonium
@@ -807,7 +809,7 @@ Boronic_acid_derivative: [BX3]([!#6])([!#6])[!#6]
 Borohydride: [BH1,BH2,BH3,BH4]
 # at least one H attached to B

-Quartary_boron: [BX4]
+Quaternary_boron: [BX4]
 # mostly borates (negative charge), in complex with Lewis-base


@@ -838,7 +840,9 @@ NH_aziridine: [NX3H1r3]1[#6r3][#6r3]1
 # toxic/reactive according to Maybridge's garbage filter

 Spiro: [D4R;$(*(@*)(@*)(@*)@*)]
-# at least two different rings can be found which are sharing just one atom 
+# at least two different rings can be found which are sharing just one atom.
+# these two rings can be connected by a third ring, so it matches also some 
+# bridged systems, like morphine

 Annelated_rings: [R;$(*(@*)(@*)@*);!$([R2;$(*(@*)(@*)(@*)@*)])]@[R;$(*(@*)(@*)@*);!$([R2;$(*(@*)(@*)(@*)@*)])]
 # two different rings sharing exactly two atoms
@@ -881,10 +885,10 @@ Sugar_pattern_2_alpha: [OX2;$([r5]1@[C@@](!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C
 Sugar_pattern_2_beta: [OX2;$([r5]1@[C@](!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C1),$([r6]1@[C@](!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C@C1)]
 # 5 or 6-membered cyclic hemi-acetal

-# Poly_sugar_1: ([OX2;$([r5]1@C@C@C(O)@C1),$([r6]1@C@C@C(O)@C(O)@C1)].[OX2;$([r5]1@C@C@C(O)@C1),$([r6]1@C@C@C(O)@C(O)@C1)])
+##Poly_sugar_1: ([OX2;$([r5]1@C@C@C(O)@C1),$([r6]1@C@C@C(O)@C(O)@C1)].[OX2;$([r5]1@C@C@C(O)@C1),$([r6]1@C@C@C(O)@C(O)@C1)])
 # pattern1 occours more than once (in same molecule, but moieties don't have to be adjacent!)

-# Poly_sugar_2: ([OX2;$([r5]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C1),$([r6]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C@C1)].[OX2;$([r5]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C1),$([r6]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C@C1)])
+##Poly_sugar_2: ([OX2;$([r5]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C1),$([r6]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C@C1)].[OX2;$([r5]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C1),$([r6]1@C(!@[OX2,NX3,SX2,FX1,ClX1,BrX1,IX1])@C@C@C@C1)])
 # pattern2 occours more than once (in same molecule, but moieties don't have to be adjacent!)


@@ -917,21 +921,21 @@ Trifluoromethyl: [FX1][CX4;!$([H0][Cl,Br,I]);!$([F][C]([F])([F])[F])]([FX1])([FX
 C_ONS_bond: [#6]~[#7,#8,#16]
 # probably all drug-like molecules have at least one O, N, or S connected to a C -> nice filter

-# Mixture: (*).(*)
+## Mixture: (*).(*)
 # two or more seperate parts, may also be salt
-# Does not work with Openbabel 2.x (no component-level grouping)
+# component-level grouping is not yet supported in Open Babel Version 2.0


 Charged: [!+0]

 Anion: [-1,-2,-3,-4,-5,-6,-7]

-Cation: [+1,+2,+3,+4,+5,+6,+7]
+Kation: [+1,+2,+3,+4,+5,+6,+7]

-# Salt: ([-1,-2,-3,-4,-5,-6,-7]).([+1,+2,+3,+4,+5,+6,+7])
+Salt: ([-1,-2,-3,-4,-5,-6,-7]).([+1,+2,+3,+4,+5,+6,+7])
 # two or more seperate components with opposite charges

-# Zwitterion: ([-1,-2,-3,-4,-5,-6,-7].[+1,+2,+3,+4,+5,+6,+7])
+##Zwitterion: ([-1,-2,-3,-4,-5,-6,-7].[+1,+2,+3,+4,+5,+6,+7])
 # both negative and positive charges somewhere within the same molecule. 

 1,3-Tautomerizable: [$([#7X2,OX1,SX1]=*[!H0;!$([a;!n])]),$([#7X3,OX2,SX2;!H0]*=*),$([#7X3,OX2,SX2;!H0]*:n)]
@@ -973,5 +977,7 @@ Chiral_center_specified: [$([*@](~*)(~*)(*)*),$([*@H](*)(*)*),$([*@](~*)(*)*),$(
 # Hits atoms with tetrahedral chirality, if chiral center is specified in the SMILES string
 # depictmach does not find oxonium, sulfonium, or sulfoxides!

-Chiral_center_unspecified: [$([*@?](~*)(~*)(*)*),$([*@?H](*)(*)*),$([*@?](~*)(*)*),$([*@?H](~*)~*)]
-# Hits atoms with tetrahedral chirality, even if chiral center is not specified in the SMILES string
+# Chiral_center_unspecified: [$([*@?](~*)(~*)(*)*),$([*@?H](*)(*)*),$([*@?](~*)(*)*),$([*@?H](~*)~*)]
+# Hits atoms with tetrahedral chirality, if chiral center is not specified in the SMILES string
+# "@?" (unspecified chirality) is not yet supported in Open Babel Version 2.0 
+ 
--- a/data/patterns.txt
+++ b/data/patterns.txt
@@ -12,7 +12,9 @@
 # SMARTS Patterns are used by finger3.cpp:PatternFP                          #
 # Format of each line is a SMARTS pattern, then optionally                   #
 #   followed by a tab character and a pattern number and/or description      #
-#   (everything after the tab will be ignored by the code                    #
+#   (everything after the tab will be ignored by the code)                   #
+# A file of this format needs the same first line as this one.               #
+# An alternative format, as in SMARTS_InteLigand.txt, can also be used       #
 #                                                                            #
 #  INCOMPLETE!! Really only useful to test the fingerprint FP3               #
 ##############################################################################
--- a/include/openbabel/fingerprint.h
+++ b/include/openbabel/fingerprint.h
@@ -53,9 +53,12 @@ public:
  virtual ~OBFingerprint(){}

  /// Sets the nth bit
-  void SetBit(std::vector<unsigned int>& vec, unsigned int n);	
+  void SetBit(std::vector<unsigned int>& vec, const unsigned int n);

-  /// Repeatedly ORs the top half with the bottom half until no smaller than nbits 
+  ///return true if the nth bit is set;
+  bool GetBit(const std::vector<unsigned int>& vec, const unsigned int n);
+
+    /// Repeatedly ORs the top half with the bottom half until no smaller than nbits 
  void Fold(std::vector<unsigned int>& vec, unsigned int nbits); 

  /// \return fingerprint in vector, which may be resized, folded to nbits (if nbits!=0)
@@ -65,6 +68,13 @@ public:
  enum FptFlag{FPT_UNIQUEBITS=1};
  virtual unsigned int Flags() { return 0;}; 

+  /// Returns a description of each bit that is set (or unset, if bSet=false)
+  virtual std::string DescribeBits(const std::  vector<unsigned int> fp, bool bSet=true)
+  {
+    std::string txt("Bit descriptions are not available for this fingerprint type");
+    return txt;
+  }
+
  /// \return the Tanimoto coefficient between two vectors (vector<unsigned int>& SeekPositions)
  static double Tanimoto(const std::vector<unsigned int>& vec1, const std::vector<unsigned int>& vec2);
  
--- a/src/fingerprint.cpp
+++ b/src/fingerprint.cpp
@@ -33,11 +33,17 @@ namespace OpenBabel

  const unsigned int OBFingerprint::bitsperint = 8 * sizeof(unsigned int);

-  void OBFingerprint::SetBit(vector<unsigned int>& vec, unsigned int n)
+  void OBFingerprint::SetBit(vector<unsigned int>& vec, const unsigned int n)
  {
    vec[n/Getbitsperint()] |= (1 << (n % Getbitsperint()));
  }

+  bool OBFingerprint::GetBit(const vector<unsigned int>& vec, const unsigned int n)
+  {
+    unsigned int word =vec[n/Getbitsperint()];
+    return (word &= (1 << (n % Getbitsperint())))!=0;
+  }
+
  ////////////////////////////////////////	
  void OBFingerprint::Fold(vector<unsigned int>& vec, unsigned int nbits)
  {
--- a/src/fingerprints/finger3.cpp
+++ b/src/fingerprints/finger3.cpp
@@ -31,65 +31,65 @@ namespace OpenBabel
 class PatternFP  : public OBFingerprint
 {
 private:
-	vector<string> smartsStrings;
+  vector<string> smartsStrings;
 protected:
-	string _patternsfile;
+  string _patternsfile;

 public:
-	PatternFP(const char* ID, const char* filename=NULL, 
-			bool IsDefault=false) : OBFingerprint(ID, IsDefault)
-	{
-		if(filename==NULL)
-			_patternsfile="patterns.txt";
-		else
-			_patternsfile = filename;
-	};
-	
-	virtual const char* Description()
-	{
-		static string desc;
+  PatternFP(const char* ID, const char* filename=NULL, 
+      bool IsDefault=false) : OBFingerprint(ID, IsDefault)
+  {
+    if(filename==NULL)
+      _patternsfile="patterns.txt";
+    else
+      _patternsfile = filename;
+  };
+  
+  virtual const char* Description()
+  {
+    static string desc;
    desc = "SMARTS patterns specified in the file " + _patternsfile;
    return (desc.c_str());
-	};
+  };

-	//Each bits represents a single substructure; no need for confirmation when substructure searching
-	virtual unsigned int Flags() { return FPT_UNIQUEBITS;}; 
+  //Each bit represents a single substructure; no need for confirmation when substructure searching
+  virtual unsigned int Flags() { return FPT_UNIQUEBITS;}; 

-	bool GetFingerprint(OBBase* pOb, vector<unsigned int>&fp, int nbits) 
-	{
-		OBMol* pmol = dynamic_cast<OBMol*>(pOb);
-		if(!pmol)
-			return false;
-		
-		//Read patterns file if it has not been done already
-		if(smartsStrings.empty())
-			ReadPatternFile(_patternsfile, smartsStrings);
+  bool GetFingerprint(OBBase* pOb, vector<unsigned int>&fp, int nbits) 
+  {
+    OBMol* pmol = dynamic_cast<OBMol*>(pOb);
+    if(!pmol)
+      return false;
+    
+    //Read patterns file if it has not been done already
+    if(smartsStrings.empty())
+      ReadPatternFile(_patternsfile, smartsStrings);

-		//Make fp size the smallest power of two to contain the patterns
-		unsigned int n=Getbitsperint();
-		while(n<smartsStrings.size())n*=2;
-		fp.resize(n/Getbitsperint());
+    //Make fp size the smallest power of two to contain the patterns
+    unsigned int n=Getbitsperint();
+    while(n<smartsStrings.size())n*=2;
+    fp.resize(n/Getbitsperint());

-		for(n=0;n<smartsStrings.size();++n)
-		{
-			OBSmartsPattern sp;
-			sp.Init(smartsStrings[n]);
-			if(sp.Match(*pmol))
-				SetBit(fp, n);
-		}
+    for(n=0;n<smartsStrings.size();++n)
+    {
+      OBSmartsPattern sp;
+      sp.Init(smartsStrings[n]);
+      if(sp.Match(*pmol))
+        SetBit(fp, n);
+    }

-		if(nbits)
-			Fold(fp, nbits);
-		return true;
-	};
+    if(nbits)
+      Fold(fp, nbits);
+    return true;
+  };

-	bool ReadPatternFile(const string& filename, vector<string>& lines)
-	{	
-		//Reads two types of file: SMARTS + comments and vice versa
-		//depending on whether the first line is #Comments after SMARTS
-		//Output strings in vector are SMARTS + comments
-		string file = filename;
-		ifstream ifs;
+  bool ReadPatternFile(const string& filename, vector<string>& lines)
+  {
+    //Reads two types of file: SMARTS + comments and vice versa
+    //depending on whether the first line is #Comments after SMARTS
+    //Output strings in vector are SMARTS + comments
+    string file = filename;
+    ifstream ifs;
 #ifdef HAVE_SSTREAM 
 	  stringstream errorMsg; 
 #else 
@@ -102,49 +102,66 @@ public:
      return false;
    }

-		if(!(ifs))
-		  {
-		    errorMsg << "Cannot open " << filename << endl; 
-		    obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obError); 
-		    return false;
-		  }
-		string smarts, formatline;
+    if(!(ifs))
+      {
+        errorMsg << "Cannot open " << filename << endl; 
+        obErrorLog.ThrowError(__FUNCTION__, errorMsg.str(), obError); 
+        return false;
+      }
+    string smarts, formatline;

-		if(!getline(ifs, formatline)) return false;
-		if(formatline=="#Comments after SMARTS")
-		{
-			while(ifs.good())
-			{
-				if( getline(ifs,smarts) 
-						&& smarts.size() > 0 
-						&& smarts[0] != '#')
-					lines.push_back(smarts); //leave the comments in
-			}
-		}
-		else
-		{
-			// Christian Laggner's format: SMARTS at end of line
-			while(ifs.good())
-			{
-				if( getline(ifs,smarts) && smarts[0]!='#')
-				{
+    if(!getline(ifs, formatline)) return false;
+    if(Trim(formatline)=="#Comments after SMARTS")
+    {
+      while(ifs.good())
+      {
+        if( getline(ifs,smarts) 
+            && Trim(smarts).size() > 0 
+            && smarts[0] != '#')
+          lines.push_back(smarts); //leave the comments in
+      }
+    }
+    else
+    {
+      // Christian Laggner's format: SMARTS at end of line
+      while(ifs.good())
+      {
+        if( getline(ifs,smarts) && smarts[0]!='#')
+        {
          string::size_type pos = smarts.find(':');
-					if(pos!=string::npos)
-					{
-						pos = smarts.find_first_not_of(" \t", pos+1);
-						if(pos!=string::npos)
-							lines.push_back(smarts.substr(pos) + ' ' + smarts.substr(0,pos));
-					}
-				}
-			}
-		}
+          if(pos!=string::npos)
+          {
+            pos = smarts.find_first_not_of(" \t", pos+1);
+            if(pos!=string::npos)
+              lines.push_back(Trim(smarts.substr(pos)) + ' ' + smarts.substr(0,pos));
+          }
+        }
+      }
+    }

    if (ifs)
      ifs.close();
-		return true;
-	}
-};
+    return true;
+  }

+
+virtual string DescribeBits(const vector<unsigned int> fp, bool bSet=true)
+{
+  stringstream ss; 
+  ss << "out of possible " << smartsStrings.size();
+  for(int i=0; i<smartsStrings.size(); ++i)
+  {
+    if(GetBit(fp, i)==bSet)
+    {
+      string::size_type pos = smartsStrings[i].find(' ');
+      if(pos!=string::npos)
+        ss << '\n' << smartsStrings[i].substr(pos+1);
+    }
+  }
+  return ss.str();
+}
+
+};
 //***********************************************
 //Make a global instance
 PatternFP thePatternFP("FP3");
--- a/src/formats/fingerprintformat.cpp
+++ b/src/formats/fingerprintformat.cpp
@@ -32,15 +32,16 @@ namespace OpenBabel

    virtual const char* Description() //required
    { return
-        "Fingerprint format\n \
-Constructs and displays fingerprints and (for multiple input objects)\n \
-the Tanimoto coefficient and whether a superstructure of the first object\n \
-Options e.g. -xfFP3 -xn128\n \
- f<id> fingerprint type\n \
- N# fold to specified number of bits, 32, 64, 128, etc.\n \
- h  hex output when multiple molecules\n \
- F  displays the available fingerprint types\n \
-";
+      "Fingerprint format\n"
+      "Constructs and displays fingerprints and (for multiple input objects)\n"
+      "the Tanimoto coefficient and whether a superstructure of the first object\n"
+      "Output options e.g. -xfFP3 -xN128\n"
+      " f<id> fingerprint type\n"
+      " N# fold to specified number of bits, 32, 64, 128, etc.\n"
+      " h  hex output when multiple molecules\n"
+      " s  describe each set bit\n"
+      " u  describe each unset bit\n"
+;
    };

    virtual unsigned int Flags(){return NOTREADABLE;};
@@ -60,13 +61,6 @@ Options e.g. -xfFP3 -xn128\n \
  bool FingerprintFormat::WriteMolecule(OBBase* pOb, OBConversion* pConv)
  {
    ostream &ofs = *pConv->GetOutStream();
-/*    if(pConv->IsOption("F"))
-      {
-        FOR_EACH(OBFingerprint, iter)
-          ofs << iter.ID() << " -- " << iter->Description() << endl;
-        return true;
-      }
-*/

    bool hexoutput=false;
    if(pConv->IsOption("h") || (pConv->GetOutputIndex()==1 && pConv->IsLast()))
@@ -111,7 +105,7 @@ Options e.g. -xfFP3 -xn128\n \
            for(;wd;wd=wd<<1)//count bits set by shifting into sign bit until word==0
              if(wd<0) ++bitsset;
          }
-        ofs  << "   " << bitsset << " bits set. "; 
+        ofs  << "   " << bitsset << " bits set "; 
      }

    if(pConv->GetOutputIndex()==1)
@@ -122,7 +116,13 @@ Options e.g. -xfFP3 -xn128\n \
          firstname=pmol->GetTitle();
        if(firstname.empty())
          firstname = "first mol";
-      }
+
+        if(pConv->IsOption("s"))
+          ofs << pFP->DescribeBits(fptvec);
+        if(pConv->IsOption("u"))
+          ofs << pFP->DescribeBits(fptvec, false);
+
+    }
    else
      {
        ofs << "   Tanimoto from " << firstname << " = " << OBFingerprint::Tanimoto(firstfp, fptvec);
--- a/src/math/spacegroup.cpp
+++ b/src/math/spacegroup.cpp
@@ -24,6 +24,7 @@ GNU General Public License for more details.
 #include <map>
 #include <set>
 #include <vector>
+#include <stdarg.h>

 using namespace std;