Enhanced stereo read/write support in SDF files. (rdkit#2022)

* add a couple test files * backup * first pass at some theory documentatin * it's a draft * Update enhanced stereochemistry documentation Adds initial target use case and caveats about the tentative nature of the current implementation. * Support read/write of molfile enhanced stereochemistry This includes reading and writing of enhanced stereochemistry from v3000 molfiles (sdf). Enhanced stereochemistry encodes the relative configuration of stereocenters, allowing representation of racemic mixtures and compounds with unknown absolute stereochemistry. It does not include: * Python wrapping * invalidation of the enhanced stereochemistry * use of enhanced stereochemistry in search * depiction of enhanced stereochemistry. * Update to reflect changes from rdkit#1971 * change names of enum elements to allow compilation in VS2017 I think it's also clearer to do things this way * Addressed most review comments. * Run missed test "testEnhancedStereoChemistry" * In tests, added size checks to group equality checks * Updated copyright statements * Deleted mol created for a test * Use perfect forwarding in RWMol::setStereoGroups() * use references for stereo groups that are checked in write and pickle * Updated stereogroup.h in hopes of fixing compilation on Windows. * clang-format * try allowing a switch to boost regex and requiring it for g++-4.8 * do a better job of that * typo * Code review comments. Updated Copyright notice. * When an atom is deleted, delete stereo groups containing it. Also updates StereoGroup toUse accessors instead of constant member attributes. This allows move of StereoGroups. * RDKit style guide * Add header required on Windows. * get the SWIG wrappers to build
msteijaert · Sep 26, 2018 · eaa44b4 · eaa44b4
1 parent ad6ce82
commit eaa44b4
Show file tree

Hide file tree

Showing 37 changed files with 691 additions and 42 deletions.
diff --git a/.travis.yml b/.travis.yml
@@ -39,7 +39,7 @@ before_install:
  - if [[ "$CONDA_PYTHON_VERSION" == "3.6" ]] ; then export PYMAJOR="3"; fi
  - if [[ "$CONDA_PYTHON_VERSION" == "2.7" ]] ; then export PYMAJOR="2"; fi
  - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then wget http://repo.continuum.io/miniconda/Miniconda$PYMAJOR-4.4.10-Linux-x86_64.sh -O miniconda.sh; fi
- - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then wget http://repo.continuum.io/miniconda/Miniconda$PYMAJOR-4.4.10-MacOSX-x86_64.sh -O miniconda.sh; fi 
+ - if [[ "$TRAVIS_OS_NAME" == "osx" ]]; then wget http://repo.continuum.io/miniconda/Miniconda$PYMAJOR-4.4.10-MacOSX-x86_64.sh -O miniconda.sh; fi
  - bash miniconda.sh -b -p $HOME/conda
  - export PATH="$HOME/conda/bin:$PATH"
  - hash -r
@@ -82,7 +82,8 @@ script:
   - cd build
   - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then  export CXX="clang++-3.9" ; fi
   - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then  export CC="clang-3.9" ; fi
-  - cmake -D Python_ADDITIONAL_VERSIONS=$CONDA_PYTHON_VERSION -D PYTHON_EXECUTABLE=$PYTHON -D PYTHON_LIBRARY=`find $PY_PREFIX -name "libpython$CONDA_PYTHON_VERSION*.so"` -D PYTHON_NUMPY_INCLUDE_PATH=$PY_SP_DIR/numpy/core/include -D BOOST_ROOT=$PY_PREFIX -D Boost_NO_SYSTEM_PATHS=ON -D RDK_BUILD_AVALON_SUPPORT=ON -D RDK_BUILD_INCHI_SUPPORT=ON -DRDK_BUILD_THREADSAFE_SSS=on -DRDK_TEST_MULTITHREADED=on -DRDK_INSTALL_STATIC_LIBS=OFF -DRDK_BUILD_SWIG_WRAPPERS=OFF -DRDK_SWIG_STATIC=OFF -DRDK_BUILD_PYTHON_WRAPPERS=OFF -DRDK_BUILD_FREESASA_SUPPORT=ON ..
+  - if [[ "$TRAVIS_OS_NAME" == "linux" ]]; then  export REGEX_EXTRA="-DRDK_USE_BOOST_REGEX=ON" ; fi
+  - cmake $REGEX_EXTRA -D Python_ADDITIONAL_VERSIONS=$CONDA_PYTHON_VERSION -D PYTHON_EXECUTABLE=$PYTHON -D PYTHON_LIBRARY=`find $PY_PREFIX -name "libpython$CONDA_PYTHON_VERSION*.so"` -D PYTHON_NUMPY_INCLUDE_PATH=$PY_SP_DIR/numpy/core/include -D BOOST_ROOT=$PY_PREFIX -D Boost_NO_SYSTEM_PATHS=ON -D RDK_BUILD_AVALON_SUPPORT=ON -D RDK_BUILD_INCHI_SUPPORT=ON -DRDK_BUILD_THREADSAFE_SSS=on -DRDK_TEST_MULTITHREADED=on -DRDK_INSTALL_STATIC_LIBS=OFF -DRDK_BUILD_SWIG_WRAPPERS=OFF -DRDK_SWIG_STATIC=OFF -DRDK_BUILD_PYTHON_WRAPPERS=OFF -DRDK_BUILD_FREESASA_SUPPORT=ON ..
   - make -j2 VERBOSE=3
   - make install
   - ls "$PY_PREFIX/lib"

diff --git a/CMakeLists.txt b/CMakeLists.txt
@@ -47,6 +47,7 @@ option(RDK_BUILD_FREESASA_SUPPORT "build the rdkit freesasa wrapper" OFF )
 option(RDK_BUILD_COORDGEN_SUPPORT "build the rdkit coordgen wrapper" ON )
 option(RDK_BUILD_MOLINTERCHANGE_SUPPORT "build in support for CommonChem molecule interchange" ON )
 option(RDK_INSTALL_DEV_COMPONENT "install libraries and headers" ON)
+option(RDK_USE_BOOST_REGEX "use boost::regex instead of std::regex (needed for systems with g++-4.8)" OFF)
 if(NOT MSVC)
   if(RDK_OPTIMIZE_NATIVE)
       set(CMAKE_CXX_FLAGS "${CMAKE_CXX_FLAGS} -mpopcnt")

diff --git a/Code/GraphMol/CMakeLists.txt b/Code/GraphMol/CMakeLists.txt
@@ -7,7 +7,7 @@ rdkit_library(GraphMol
               AtomIterators.cpp BondIterators.cpp Aromaticity.cpp Kekulize.cpp
               MolDiscriminators.cpp ConjugHybrid.cpp AddHs.cpp
               Matrices.cpp Chirality.cpp RingInfo.cpp Conformer.cpp
-              Renumber.cpp AdjustQuery.cpp Resonance.cpp
+              Renumber.cpp AdjustQuery.cpp Resonance.cpp StereoGroup.cpp
               new_canon.cpp
               SHARED
               LINK_LIBRARIES RDGeometryLib RDGeneral
@@ -39,6 +39,7 @@ rdkit_headers(Atom.h
               ROMol.h
               RWMol.h
               SanitException.h
+              StereoGroup.h
               MonomerInfo.h
               new_canon.h
               MolBundle.h

diff --git a/Code/GraphMol/FileParsers/CMakeLists.txt b/Code/GraphMol/FileParsers/CMakeLists.txt
@@ -10,6 +10,12 @@ else()
     set (maeparser_var "")
 endif()
 
+if(RDK_USE_BOOST_REGEX)
+  add_definitions(-DRDKIT_USE_BOOST_REGEX)
+  find_package(Boost 1.58.0 COMPONENTS regex REQUIRED)
+  set(regex_lib Boost::regex)
+endif()
+
 rdkit_library(FileParsers
               Mol2FileParser.cpp
               MolFileParser.cpp MolFileStereochem.cpp MolFileWriter.cpp
@@ -22,7 +28,7 @@ rdkit_library(FileParsers
               ProximityBonds.cpp
               SequenceParsers.cpp SequenceWriters.cpp
               SVGParser.cpp
-              LINK_LIBRARIES Depictor SmilesParse GraphMol ${maeparser_var})
+              LINK_LIBRARIES Depictor SmilesParse GraphMol ${maeparser_var} ${regex_lib})
 
 rdkit_headers(FileParsers.h
               FileParserUtils.h
@@ -59,4 +65,7 @@ rdkit_test(testMol2ToMol testMol2ToMol.cpp LINK_LIBRARIES FileParsers SmilesPars
 
 rdkit_test(testSequence testSequence.cpp LINK_LIBRARIES FileParsers SmilesParse GraphMol RDGeneral RDGeometryLib )
 
+rdkit_test(testExtendedStereoParsing testExtendedStereoParsing.cpp
+           LINK_LIBRARIES FileParsers Depictor GraphMol RDGeneral RDGeometryLib )
+
 rdkit_catch_test(fileParsersCatchTest catch_tests.cpp LINK_LIBRARIES FileParsers SmilesParse GraphMol RDGeneral RDGeometryLib )
diff --git a/Code/GraphMol/FileParsers/MolFileParser.cpp b/Code/GraphMol/FileParsers/MolFileParser.cpp
@@ -20,6 +20,7 @@
 
 #include <GraphMol/SmilesParse/SmilesParse.h>
 #include <GraphMol/RDKitQueries.h>
+#include <GraphMol/StereoGroup.h>
 #include <RDGeneral/StreamOps.h>
 #include <RDGeneral/RDLog.h>
 
@@ -29,9 +30,21 @@
 #include <RDGeneral/LocaleSwitcher.h>
 #include <typeinfo>
 #include <exception>
+#ifdef RDKIT_USE_BOOST_REGEX
+#include <boost/regex.hpp>
+using boost::regex;
+using boost::regex_match;
+using boost::smatch;
+#else
+#include <regex>
+using std::regex;
+using std::regex_match;
+using std::smatch;
+#endif
 #include <sstream>
 #include <locale>
 #include <stdlib.h>
+#include <cstdio>
 
 namespace RDKit {
 class MolFileUnhandledFeatureException : public std::exception {
@@ -139,7 +152,7 @@ void completeQueryAndChildren(ATOM_EQUALS_QUERY *query, Atom *tgt,
                              magicVal);
   }
 }
-void CompleteMolQueries(RWMol *mol, int magicVal = 0xDEADBEEF) {
+void completeMolQueries(RWMol *mol, int magicVal = 0xDEADBEEF) {
   for (ROMol::AtomIterator ai = mol->beginAtoms(); ai != mol->endAtoms();
        ++ai) {
     if ((*ai)->hasQuery()) {
@@ -150,6 +163,71 @@ void CompleteMolQueries(RWMol *mol, int magicVal = 0xDEADBEEF) {
   }
 }
 
+bool startsWith(const std::string &haystack, const char *needle, size_t size) {
+  return haystack.compare(0u, size, needle, size) == 0;
+}
+
+//! parse a collection block to find enhanced stereo groups
+std::string parseEnhancedStereo(std::istream *inStream, unsigned int &line,
+                                RWMol *mol) {
+  // Lines like (absolute, relative, racemic):
+  // M  V30 MDLV30/STEABS ATOMS=(2 2 3)
+  // M  V30 MDLV30/STEREL1 ATOMS=(1 12)
+  // M  V30 MDLV30/STERAC1 ATOMS=(1 12)
+  const regex stereo_label(
+      R"regex(MDLV30/STE(...)[0-9]* +ATOMS=\(([0-9]+) +(.*)\))regex");
+
+  smatch match;
+  std::vector<StereoGroup> groups;
+
+  // Read the collection until the end
+  auto tempStr = getV3000Line(inStream, line);
+  boost::to_upper(tempStr);
+  while (!startsWith(tempStr, "END", 3)) {
+    // If this line in the collection is part of a stereo group
+    if (regex_match(tempStr, match, stereo_label)) {
+      StereoGroupType grouptype = RDKit::StereoGroupType::STEREO_ABSOLUTE;
+
+      if (match[1] == "ABS") {
+        grouptype = RDKit::StereoGroupType::STEREO_ABSOLUTE;
+      } else if (match[1] == "REL") {
+        grouptype = RDKit::StereoGroupType::STEREO_OR;
+      } else if (match[1] == "RAC") {
+        grouptype = RDKit::StereoGroupType::STEREO_AND;
+      } else {
+        std::ostringstream errout;
+        errout << "Unrecognized stereogroup type : '" << tempStr << "' on line"
+               << line;
+        throw FileParseException(errout.str());
+      }
+
+      const unsigned int count = FileParserUtils::toInt(match[2], true);
+      std::vector<Atom *> atoms;
+      std::stringstream ss(match[3]);
+      unsigned int index;
+      for (size_t i = 0; i < count; ++i) {
+        ss >> index;
+        // atoms are 1 indexed in molfiles
+        atoms.push_back(mol->getAtomWithIdx(index - 1));
+      }
+      groups.emplace_back(grouptype, std::move(atoms));
+    } else {
+      // skip collection types we don't know how to read. Only one documented
+      // is MDLV30/HILITE
+      BOOST_LOG(rdWarningLog) << "Skipping unrecognized collection type at "
+                                 "line "
+                              << line << ": " << tempStr << std::endl;
+    }
+    tempStr = getV3000Line(inStream, line);
+  }
+
+  if (!groups.empty()) {
+    mol->setStereoGroups(std::move(groups));
+  }
+  tempStr = getV3000Line(inStream, line);
+  return tempStr;
+}
+
 //*************************************
 //
 // Every effort has been made to adhere to MDL's standard
@@ -2321,15 +2399,17 @@ bool ParseV3000CTAB(std::istream *inStream, unsigned int &line, RWMol *mol,
   }
 
   while (tempStr.length() > 5 && tempStr.substr(0, 5) == "BEGIN") {
-    // skip blocks we don't know how to read
-    BOOST_LOG(rdWarningLog)
-        << "skipping block at line " << line << ": " << tempStr << std::endl;
-    tempStr = getV3000Line(inStream, line);
-
-    while (tempStr.length() < 3 || tempStr.substr(0, 3) != "END") {
+    if (tempStr.length() > 15 && tempStr.substr(6, 10) == "COLLECTION") {
+      tempStr = parseEnhancedStereo(inStream, line, mol);
+    } else {
+      // skip blocks we don't know how to read
+      BOOST_LOG(rdWarningLog)
+          << "skipping block at line " << line << ": " << tempStr << std::endl;
+      while (tempStr.length() < 3 || tempStr.substr(0, 3) != "END") {
+        tempStr = getV3000Line(inStream, line);
+      }
       tempStr = getV3000Line(inStream, line);
     }
-    tempStr = getV3000Line(inStream, line);
   }
 
   boost::to_upper(tempStr);
@@ -2658,7 +2738,7 @@ RWMol *MolDataStreamToMol(std::istream *inStream, unsigned int &line,
 
     if (res->hasProp(common_properties::_NeedsQueryScan)) {
       res->clearProp(common_properties::_NeedsQueryScan);
-      CompleteMolQueries(res);
+      completeMolQueries(res);
     }
   }
   return res;

diff --git a/Code/GraphMol/FileParsers/MolFileWriter.cpp b/Code/GraphMol/FileParsers/MolFileWriter.cpp
@@ -111,13 +111,14 @@ int getQueryBondSymbol(const Bond *bond) {
           }
         }
       }
-    } else if( qry->getDescription() == "SingleOrAromaticBond" && !qry->getNegation()) {
+    } else if (qry->getDescription() == "SingleOrAromaticBond" &&
+               !qry->getNegation()) {
       res = 6;
     }
   }
   return res;
 }
-}
+}  // namespace
 
 const std::string GetMolFileChargeInfo(const RWMol &mol) {
   std::stringstream res;
@@ -276,9 +277,8 @@ const std::string GetMolFileQueryInfo(const RWMol &mol) {
       wrote_query = true;
     }
     std::string molFileValue;
-    if (!wrote_query &&
-        (*atomIt)->getPropIfPresent(common_properties::molFileValue,
-                                    molFileValue))
+    if (!wrote_query && (*atomIt)->getPropIfPresent(
+                            common_properties::molFileValue, molFileValue))
       ss << "V  " << std::setw(3) << (*atomIt)->getIdx() + 1 << " "
          << molFileValue << std::endl;
   }
@@ -510,7 +510,7 @@ unsigned int getAtomParityFlag(const Atom *atom, const Conformer *conf) {
   }
   return 0;
 }
-}
+}  // namespace
 
 bool hasNonDefaultValence(const Atom *atom) {
   if (atom->getNumRadicalElectrons() != 0) return true;
@@ -656,7 +656,7 @@ class RequiresV3000Exception : public std::runtime_error {
   explicit RequiresV3000Exception()
       : std::runtime_error("RequiresV3000Exception"){};
 };
-}
+}  // namespace
 
 int BondGetMolFileSymbol(const Bond *bond) {
   PRECONDITION(bond, "");
@@ -778,7 +778,7 @@ void GetMolFileBondStereoInfo(const Bond *bond, const INT_MAP_INT &wedgeBonds,
           boost::tie(beg, end) =
               bond->getOwningMol().getAtomBonds(bond->getBeginAtom());
           while (beg != end && !nbrHasDir) {
-            const Bond* nbrBond = bond->getOwningMol()[*beg];
+            const Bond *nbrBond = bond->getOwningMol()[*beg];
             if (nbrBond->getBondType() == Bond::SINGLE &&
                 (nbrBond->getBondDir() == Bond::ENDUPRIGHT ||
                  nbrBond->getBondDir() == Bond::ENDDOWNRIGHT)) {
@@ -789,7 +789,7 @@ void GetMolFileBondStereoInfo(const Bond *bond, const INT_MAP_INT &wedgeBonds,
           boost::tie(beg, end) =
               bond->getOwningMol().getAtomBonds(bond->getEndAtom());
           while (beg != end && !nbrHasDir) {
-            const Bond* nbrBond = bond->getOwningMol()[*beg];
+            const Bond *nbrBond = bond->getOwningMol()[*beg];
             if (nbrBond->getBondType() == Bond::SINGLE &&
                 (nbrBond->getBondDir() == Bond::ENDUPRIGHT ||
                  nbrBond->getBondDir() == Bond::ENDDOWNRIGHT)) {
@@ -1009,6 +1009,42 @@ const std::string GetV3000MolFileBondLine(const Bond *bond,
   return ss.str();
 }
 
+void appendEnhancedStereoGroups(std::string &res, const RWMol &tmol) {
+  unsigned or_count = 1u, and_count = 1u;
+  auto &stereo_groups = tmol.getStereoGroups();
+  if (!stereo_groups.empty()) {
+    res += "M  V30 BEGIN COLLECTION\n";
+    for (auto &&group : stereo_groups) {
+      res += "M  V30 MDLV30/";
+      switch (group.getGroupType()) {
+        case RDKit::StereoGroupType::STEREO_ABSOLUTE:
+          res += "STEABS";
+          break;
+        case RDKit::StereoGroupType::STEREO_OR:
+          res += "STEREL";
+          res += boost::lexical_cast<std::string>(or_count);
+          ++or_count;
+          break;
+        case RDKit::StereoGroupType::STEREO_AND:
+          res += "STERAC";
+          res += boost::lexical_cast<std::string>(and_count);
+          ++and_count;
+          break;
+      }
+      res += " ATOMS=(";
+      auto& atoms = group.getAtoms();
+      res += boost::lexical_cast<std::string>(atoms.size());
+      for (auto &&atom : atoms) {
+        res += ' ';
+        // atoms are 1 indexed in molfiles
+        res += boost::lexical_cast<std::string>(atom->getIdx() + 1);
+      }
+      res += ")\n";
+    }
+    res += "M  V30 END COLLECTION\n";
+  }
+}
+
 //------------------------------------------------
 //
 //  gets a mol block as a string
@@ -1073,7 +1109,8 @@ std::string outputMolToMolBlock(const RWMol &tmol, int confId,
   if (forceV3000)
     isV3000 = true;
   else
-    isV3000 = (nAtoms > 999) || (nBonds > 999);
+    isV3000 =
+        (nAtoms > 999) || (nBonds > 999) || !tmol.getStereoGroups().empty();
 
   // the counts line:
   std::stringstream ss;
@@ -1157,6 +1194,8 @@ std::string outputMolToMolBlock(const RWMol &tmol, int confId,
       }
       res += "M  V30 END BOND\n";
     }
+    appendEnhancedStereoGroups(res, tmol);
+
     res += "M  V30 END CTAB\n";
   }
   res += "M  END\n";
@@ -1219,4 +1258,4 @@ void MolToMolFile(const ROMol &mol, const std::string &fName,
   *outStream << outString;
   delete outStream;
 }
-}
+}  // namespace RDKit