aizynthfinder.search.dfpn package

Submodules

aizynthfinder.search.dfpn.nodes module

Module containing a classes representation various tree nodes

class aizynthfinder.search.dfpn.nodes.MoleculeNode(mol, config, owner, parent=None)

Bases: _SuperNode

An OR node representing a molecule

Variables:

  • expandable – if True, this node is part of the frontier

  • mol – the molecule represented by the node

  • in_stock – if True the molecule is in stock and hence should not be expanded

  • parent – the parent of the node

  • pn – the proof number

  • dn – the disproof number

  • pn_threshold – the threshold for proof number

  • dn_threshold – the threshold for disproof number

Parameters:
classmethod create_root(smiles, config, owner)

Create a root node for a tree using a SMILES.

Parameters:

  • smiles (str) – the SMILES representation of the root state

  • config (Configuration) – settings of the tree search algorithm

  • owner (SearchTree)

Returns:

the created node

Return type:

MoleculeNode

property prop: StrDict

Dictionary with publicly exposed properties

expand()

Expand the molecule by utilising an expansion policy

Return type:

None

promising_child()

Find and return the most promising child for exploration Updates the thresholds on that child

Return type:

Optional[ReactionNode]

update()

Update the proof and disproof numbers

Return type:

None

class aizynthfinder.search.dfpn.nodes.ReactionNode(reaction, config, owner, parent)

Bases: _SuperNode

An AND node representing a reaction

Variables:

  • parent – the parent of the node

  • reaction – the reaction represented by the node

  • pn – the proof number

  • dn – the disproof number

  • pn_threshold – the threshold for proof number

  • dn_threshold – the threshold for disproof number

  • expandable – if the node is expandable

Parameters:
property children: List[MoleculeNode]

Gives the molecule children nodes

property prop: StrDict

Dictionary with publicly exposed properties

property proven: bool

Return if the node is proven

property disproven: bool

Return if the node is disproven

expand()

Expand the node by creating nodes for each reactant

Return type:

None

promising_child()

Find and return the most promising child for exploration Updates the thresholds on that child

Return type:

Optional[MoleculeNode]

update()

Update the proof and disproof numbers

Return type:

None

aizynthfinder.search.dfpn.search_tree module

Module containing a class that holds the tree search

class aizynthfinder.search.dfpn.search_tree.SearchTree(config, root_smiles=None)

Bases: AndOrSearchTreeBase

Encapsulation of the Depth-First Proof-Number (DFPN) search algorithm.

This algorithm does not support:

  1. Filter policy

  2. Serialization and deserialization

Variables:

  • config – settings of the tree search algorithm

  • root – the root node

Parameters:

  • config (Configuration) – settings of the tree search algorithm

  • root_smiles (Optional[str]) – the root will be set to a node representing this molecule, defaults to None

property mol_nodes: Sequence[MoleculeNode]

Return the molecule nodes of the tree

one_iteration()

Perform one iteration of expansion.

If possible expand the frontier node twice, i.e. expanding an OR node and then and AND node. If frontier not expandable step up in the tree and find a new frontier to expand.

If a solution is found, mask that tree for exploration and start over.

Raises:

StopIteration – if the search should be pre-maturely terminated

Returns:

if a solution was found

Return type:

bool

routes()

Extracts and returns routes from the AND/OR tree

Returns:

the routes

Return type:

List[ReactionTree]

Module contents

Sub-package containing DFPN routines