Merge for version 3.4.0

CHANGELOG.md

@@ -1,5 +1,16 @@
 # CHANGELOG
 
+## Version 3.4.0 2022-04-28
+
+### Features
+
+- Gracefully fail predictions with aizynthcli if target is unsanitizable (Github issue 66)
+
+### Trivial changes
+- Update version of Sphinx, scikit-learn and Black
+- Remove usage of depracted Pandas function append (Github issue 63)
+- Correct documentation of aizynthcli (Github issue 67)
+
 ## Version 3.3.1 2022-03-09
 
 ### Trivial changes

README.md

@@ -18,7 +18,7 @@ Before you begin, ensure you have met the following requirements:
 
 * Linux, Windows or macOS platforms are supported - as long as the dependencies are supported on these platforms.
 
-* You have installed [anaconda](https://www.anaconda.com/) or [miniconda](https://docs.conda.io/en/latest/miniconda.html) with python 3.6 - 3.8
+* You have installed [anaconda](https://www.anaconda.com/) or [miniconda](https://docs.conda.io/en/latest/miniconda.html) with python 3.6 - 3.9
 
 The tool has been developed on a Linux platform, but the software has been tested on Windows 10 and macOS Catalina.
 
@@ -62,7 +62,7 @@ First, install these conda packages
 
 Secondly, install the ``aizynthfinder`` package
 
-    python -m pip install https://github.com/MolecularAI/aizynthfinder/archive/v3.3.1.tar.gz
+    python -m pip install https://github.com/MolecularAI/aizynthfinder/archive/v3.4.0.tar.gz
 
 
 if you want to install the latest version

aizynthfinder/aizynthfinder.py

@@ -20,6 +20,7 @@
 )
 from aizynthfinder.chem import Molecule, TreeMolecule, FixedRetroReaction
 from aizynthfinder.search.andor_trees import AndOrSearchTreeBase
+from aizynthfinder.utils.exceptions import MoleculeException
 
 if TYPE_CHECKING:
     from aizynthfinder.utils.type_utils import (
@@ -148,6 +149,11 @@ def prepare_tree(self) -> None:
         if not self.target_mol:
             raise ValueError("No target molecule set")
 
+        try:
+            self.target_mol.sanitize()
+        except MoleculeException:
+            raise ValueError("Target molecule unsanitizable")
+
         self.stock.reset_exclusion_list()
         if self.config.exclude_target_from_stock and self.target_mol in self.stock:
             self.stock.exclude(self.target_mol)

aizynthfinder/analysis/routes.py

@@ -133,7 +133,7 @@ def cluster(
         n_clusters: int,
         max_clusters: int = 5,
         distances_model: str = "ted",
-        **kwargs: Any
+        **kwargs: Any,
     ) -> np.ndarray:
         """
         Cluster the route collection into a number of clusters.

aizynthfinder/analysis/utils.py

@@ -37,6 +37,7 @@ class RouteSelectionArguments:
     If `return_all` is True, it will return all solved routes if there is at least one is solved, otherwise
     the `nmin` and `nmax` will be used.
     """
+
     nmin: int = 5
     nmax: int = 25
     return_all: bool = False

aizynthfinder/chem/mol.py

@@ -100,7 +100,7 @@ def inchi_key(self) -> str:
 
     @property
     def weight(self) -> float:
-        """ Return the exact molecular weight of the molecule """
+        """Return the exact molecular weight of the molecule"""
         self.sanitize(raise_exception=False)
         return Descriptors.ExactMolWt(self.rd_mol)
 

aizynthfinder/chem/reaction.py

@@ -332,7 +332,7 @@ def __str__(self) -> str:
 
     @property
     def rd_reaction(self) -> RdReaction:
-        """ Return the RDKit reaction created from the SMART """
+        """Return the RDKit reaction created from the SMART"""
         if self._rd_reaction is None:
             self._rd_reaction = AllChem.ReactionFromSmarts(self.smarts)
         return self._rd_reaction

aizynthfinder/context/cost/collection.py

@@ -17,7 +17,7 @@
 
 
 class MoleculeCost(ContextCollection):
-    """ Collection of molecular cost objects """
+    """Collection of molecular cost objects"""
 
     _single_selection: bool = True
     _collection_name: str = "molecule cost"

aizynthfinder/interfaces/aizynthcli.py

@@ -141,7 +141,11 @@ def _process_single_smiles(
 ) -> None:
     output_name = output_name or "trees.json"
     finder.target_smiles = smiles
-    finder.prepare_tree()
+    try:
+        finder.prepare_tree()
+    except ValueError as err:
+        print(f"Failed to setup search due to: '{str(err).lower()}'")
+        return
     finder.tree_search(show_progress=True)
     finder.build_routes()
 
@@ -179,7 +183,11 @@ def _process_multi_smiles(
     results = defaultdict(list)
     for smi in smiles:
         finder.target_smiles = smi
-        finder.prepare_tree()
+        try:
+            finder.prepare_tree()
+        except ValueError as err:
+            print(f"Failed to setup search for {smi} due to: '{str(err).lower()}'")
+            continue
         search_time = finder.tree_search()
         finder.build_routes()
         stats = finder.extract_statistics()

aizynthfinder/search/andor_trees.py

@@ -17,39 +17,39 @@
 
 
 class TreeNodeMixin:
-    """ A mixin class for node in a tree """
+    """A mixin class for node in a tree"""
 
     @property
     def prop(self) -> StrDict:
-        """ Dictionary with publicly exposed properties """
+        """Dictionary with publicly exposed properties"""
         return {}
 
     @property
     def children(self) -> List["TreeNodeMixin"]:
-        """ List of children nodes """
+        """List of children nodes"""
         return []
 
 
 class AndOrSearchTreeBase(abc.ABC):
-    """ A base class for a search tree based on an AND/OR structure """
+    """A base class for a search tree based on an AND/OR structure"""
 
     def __init__(self, config: Configuration, root_smiles: str = None) -> None:
         self.config = config
         self._root_smiles = root_smiles
 
     @property
     def mol_nodes(self) -> List[TreeNodeMixin]:
-        """ Return the molecule nodes of the tree """
+        """Return the molecule nodes of the tree"""
         return []
 
     @abc.abstractmethod
     def one_iteration(self) -> bool:  # pylint: disable=no-self-use
-        """ Perform one iteration of the search """
+        """Perform one iteration of the search"""
         return False
 
     @abc.abstractmethod
     def routes(self) -> List[ReactionTree]:  # pylint: disable=no-self-use
-        """ Return the routes of the tree """
+        """Return the routes of the tree"""
         return []
 
 
@@ -150,7 +150,7 @@ def __init__(self, root: TreeNodeMixin = None) -> None:
 
     @property
     def first_reaction(self) -> TreeNodeMixin:
-        """ Return the first reaction or raise an exception """
+        """Return the first reaction or raise an exception"""
         assert self._first_reaction is not None
         return self._first_reaction
 

aizynthfinder/search/breadth_first/nodes.py

@@ -91,7 +91,7 @@ def from_dict(
 
     @property  # type: ignore
     def children(self) -> List[ReactionNode]:  # type: ignore
-        """ Gives the reaction children nodes """
+        """Gives the reaction children nodes"""
         return self._children
 
     @children.setter
@@ -220,7 +220,7 @@ def from_dict(
 
     @property  # type: ignore
     def children(self) -> List[MoleculeNode]:  # type: ignore
-        """ Gives the molecule children nodes """
+        """Gives the molecule children nodes"""
         return self._children
 
     @children.setter

aizynthfinder/search/breadth_first/search_tree.py

@@ -76,7 +76,7 @@ def _find_mol_nodes(node):
 
     @property
     def mol_nodes(self) -> Sequence[MoleculeNode]:  # type: ignore
-        """ Return the molecule nodes of the tree """
+        """Return the molecule nodes of the tree"""
         return self._mol_nodes
 
     def one_iteration(self) -> bool:

aizynthfinder/search/dfpn/nodes.py

@@ -30,32 +30,32 @@ def __init__(self) -> None:
 
     @property  # type: ignore
     def children(self) -> List[ReactionNode]:  # type: ignore
-        """ Gives the reaction children nodes """
+        """Gives the reaction children nodes"""
         return self._children  # type: ignore
 
     @property
     def closed(self) -> bool:
-        """ Return if the node is proven or disproven """
+        """Return if the node is proven or disproven"""
         return self.proven or self.disproven
 
     @property
     def proven(self) -> bool:
-        """ Return if the node is proven"""
+        """Return if the node is proven"""
         return self.pn == 0
 
     @property
     def disproven(self) -> bool:
-        """ Return if the node is disproven"""
+        """Return if the node is disproven"""
         return self.dn == 0
 
     def explorable(self) -> bool:
-        """ Return if the node can be explored by the search algorithm"""
+        """Return if the node can be explored by the search algorithm"""
         return not (
             self.closed or self.pn > self.pn_threshold or self.dn > self.dn_threshold
         )
 
     def reset(self) -> None:
-        """ Reset the thresholds """
+        """Reset the thresholds"""
         if self.closed or self.expandable:
             return
         for child in self._children:
@@ -65,7 +65,7 @@ def reset(self) -> None:
         self.dn_threshold = BIG_INT
 
     def update(self) -> None:
-        """ Update the proof and disproof numbers """
+        """Update the proof and disproof numbers"""
         raise NotImplementedError("Implement a child class")
 
     def _set_disproven(self) -> None:
@@ -139,7 +139,7 @@ def prop(self) -> StrDict:
         return {"solved": self.proven, "mol": self.mol}
 
     def expand(self) -> None:
-        """ Expand the molecule by utilising an expansion policy """
+        """Expand the molecule by utilising an expansion policy"""
         self.expandable = False
         reactions, priors = self._config.expansion_policy([self.mol])
         self.tree.profiling["expansion_calls"] += 1
@@ -194,7 +194,7 @@ def promising_child(self) -> Optional[ReactionNode]:
         return best_child
 
     def update(self) -> None:
-        """ Update the proof and disproof numbers """
+        """Update the proof and disproof numbers"""
         func = all if self.parent is None else any
         if func(child.proven for child in self._children):
             self._set_proven()
@@ -277,7 +277,7 @@ def __init__(
 
     @property  # type: ignore
     def children(self) -> List[MoleculeNode]:  # type: ignore
-        """ Gives the molecule children nodes """
+        """Gives the molecule children nodes"""
         return self._children  # type: ignore
 
     @property
@@ -286,7 +286,7 @@ def prop(self) -> StrDict:
 
     @property
     def proven(self) -> bool:
-        """ Return if the node is proven """
+        """Return if the node is proven"""
         if self.expandable:
             return False
         if self.pn == 0:
@@ -295,15 +295,15 @@ def proven(self) -> bool:
 
     @property
     def disproven(self) -> bool:
-        """ Return if the node is disproven """
+        """Return if the node is disproven"""
         if self.expandable:
             return False
         if self.dn == 0:
             return True
         return any(child.disproven for child in self._children)
 
     def expand(self) -> None:
-        """ Expand the node by creating nodes for each reactant """
+        """Expand the node by creating nodes for each reactant"""
         self.expandable = False
         reactants = self.reaction.reactants[self.reaction.index]
         self._children = [
@@ -331,7 +331,7 @@ def promising_child(self) -> Optional[MoleculeNode]:
         return best_child
 
     def update(self) -> None:
-        """ Update the proof and disproof numbers"""
+        """Update the proof and disproof numbers"""
         if all(child.proven for child in self._children):
             self._set_proven()
             return

aizynthfinder/search/dfpn/search_tree.py

@@ -55,7 +55,7 @@ def __init__(self, config: Configuration, root_smiles: str = None) -> None:
 
     @property
     def mol_nodes(self) -> Sequence[MoleculeNode]:  # type: ignore
-        """ Return the molecule nodes of the tree """
+        """Return the molecule nodes of the tree"""
         return self._mol_nodes
 
     def one_iteration(self) -> bool:

aizynthfinder/search/mcts/node.py

@@ -167,7 +167,7 @@ def parent(self) -> Optional["MctsNode"]:
 
     @property
     def state(self) -> MctsState:
-        """ Return the underlying state of the node """
+        """Return the underlying state of the node"""
         return self._state
 
     def actions_to(self) -> List[RetroReaction]:

aizynthfinder/search/mcts/search.py

@@ -99,7 +99,7 @@ def add_node(node):
         return self._graph
 
     def nodes(self) -> List[MctsNode]:
-        """ Return all the nodes in the search tree """
+        """Return all the nodes in the search tree"""
         return list(self.graph())
 
     def one_iteration(self) -> bool:

aizynthfinder/search/retrostar/nodes.py

@@ -101,7 +101,7 @@ def from_dict(
 
     @property  # type: ignore
     def children(self) -> List[ReactionNode]:  # type: ignore
-        """ Gives the reaction children nodes """
+        """Gives the reaction children nodes"""
         return self._children
 
     @children.setter
@@ -301,7 +301,7 @@ def from_dict(
 
     @property  # type: ignore
     def children(self) -> List[MoleculeNode]:  # type: ignore
-        """ Gives the molecule children nodes """
+        """Gives the molecule children nodes"""
         return self._children
 
     @children.setter

aizynthfinder/search/retrostar/search_tree.py

@@ -78,7 +78,7 @@ def _find_mol_nodes(node):
 
     @property
     def mol_nodes(self) -> Sequence[MoleculeNode]:  # type: ignore
-        """ Return the molecule nodes of the tree """
+        """Return the molecule nodes of the tree"""
         return self._mol_nodes
 
     def one_iteration(self) -> bool:

aizynthfinder/training/make_false_products.py

@@ -257,7 +257,7 @@ def main(optional_args: Optional[Sequence[str]] = None) -> None:
     errors: List[str] = []
     for new_df in methods[selected_method](library, config, errors):
         if new_df is not None:
-            false_lib = false_lib.append(new_df, sort=False)
+            false_lib = pd.concat([false_lib, new_df])
         progress_bar.update(1)
     progress_bar.close()
 

aizynthfinder/training/preprocess_filter.py

@@ -46,7 +46,7 @@ def main(optional_args: Optional[Sequence[str]] = None) -> None:
         sep=config["csv_sep"],
     )
     false_dataset["true_product"] = 0
-    dataset = true_dataset.append(false_dataset, sort=False)
+    dataset = pd.concat([true_dataset, false_dataset])
 
     if config["reaction_smiles_column"]:
         dataset = split_reaction_smiles(dataset, config)