"""
The `Molecule` class is a wrapper around a biopython structure and a core part
of biobuild functionality. It provides a convenient interface to molecular structures
and their properties, such as atoms, bonds, residues, chains, etc.
.. note::
To help with identifying individual atoms, residues, etc. biobuild uses a different identification scheme than biopython does.
Therefore biobuild comes with its own child classes of the biopython classes
that are used to represent the structure. These classes are called `Atom`, `Residue`, `Chain`, etc.
and can be used as drop-in replacements for the biopython classes and should not break any existing code.
However, in case any incompatibility is observed anyway, the classes are equipped with a `to_biopython` method
that will remove the biobuild overhead and return pure biopython objects (this is not an in-place operation, however,
and will return a new object).
Making Molecules
================
The easiest way to make a new molecule is to use the toplevel `molecule` function, which will automatically
try to detect the type of user provided input and generate a molecule from it. Currently supported inputs are:
- A biopython structure
- A PDB id
- A PDB file
- A CIF file
- A SMILES string
- An InChI string
- An IUPAC name or abbreviation, or any name that matches a known compound synonym that is associated with the PubChem database
.. code-block:: python
from biobuild import molecule
my_glucose = molecule("GLC") # use the PDB id
# or
my_glucose = molecule("GLC.pdb") # use the PDB file
# or
my_glucose = molecule("alpha-d-glucose") # use the name
# ...
Since the `molecule` function is a try-and-error function it is convenient but not the most efficient. Hence, the `Molecule` class
offers already a number of convenient methods to easily generate molecules directly from specified data sources. Available methods are:
- `Molecule.from_pdb` to generate a molecule from a PDB file
- `Molecule.from_cif` to generate a molecule from a CIF file
- `Molecule.from_smiles` to generate a molecule from a SMILES string
- `Molecule.from_pubchem` to generate a molecule from a PubChem entry
- `Molecule.from_compound` to generate a molecule from a PDBECompounds entry
- `Molecule.from_rdkit` to generate a molecule from an RDKit molecule object
Hence, if we know that "glucose" is already available in our local PDBECompounds database, we can generate the molecule also as follows:
.. code-block:: python
from biobuild import Molecule
my_glucose = Molecule.from_compound("GLC") # use the PDB id
The quickest way to query the local PDBECompounds database is to use the `PDB Id` of the desired compounds.
However, the `from_compound` accepts other inputs as well. The database is queried using the `by` parameter, which can be one of the following:
- "id" for the PDB id (default)
- "name" for the name of the compound (must match any known synonym of the iupac name)
- "formula" for the chemical formula (usually ambiguous and will therefore often raise an error)
- "smiles" for the SMILES string (also accepts InChI)
.. code-block:: python
# create a new glucose molecule
glc = Molecule.from_compound("alpha-d-glucose", by="name") # use the name
Saving Molecules
----------------
If a molecule is created from a large PDB file and has undergone a lot of preprocessing, it may be useful to `save` the molecule to a pickle file
which can be loaded again later to avoid repeated preprocessing. This can be done using the `save` and `load` methods.
.. code-block:: python
# save the molecule to a pickle file
my_molecule.save("my_molecule.pkl")
# load the molecule from the pickle file
my_molecule = Molecule.load("my_molecule.pkl")
Modifying Molecules
===================
Once a molecule is created, it can be modified in a number of ways. The most common modifications are
- adding bonds (because when loading from a PDB file, bonds are not inferred automatically!)
- adding additional residues and atoms
- removing residues and atoms
- adjusting labelling (e.g. changing the chain names and residue seqids)
Adding bonds
------------
To add a bond between two atoms, we use the `add_bond` method, which expects two arguments for the two connected atoms.
The atoms can be specified by their `full_id` tuple, their `id` string, their `serial_number` (always starting at 1) or directly (the `biopython.Atom` object).
.. code-block:: python
glc = Molecule.from_compound("GLC")
# add a bond between the first and second atom
glc.add_bond(1, 2)
# and also add a bond between "O1" and "HO1" atoms
glc.add_bond("O1", "HO1")
Already for small molecules such as glucose with only 24 atoms, it would be very tedious to add all bonds manually. Good thing that the molecules
created using `from_compound` or `from_pdb` already contain all the default bonds!
However, in case the bonds are missing, or the PDB file did not specify any to begin with, the `Molecule` class offers two methods: `apply_standard_bonds` and `infer_bonds`.
The former uses reference connectivity information from the PDBECompounds database or CHARMMTopology to add all bonds that are known for the compound (if it exists in the database).
The latter will use a simple distance-based approach to infer bonds between atoms that are closer than a specified threshold (default: 1.6 Å), which can be restricted
further to a min-max window.
.. code-block:: python
# add all standard bonds for Glucose
glc.apply_standard_bonds()
# add all bonds that are closer than 1.6 Å
glc.infer_bonds(bond_length=1.6)
# add all bonds that are closer than 1.6 Å, but not closer than 1.0 Å
glc.infer_bonds(bond_length=(1.0, 1.6))
.. note::
By default `infer_bonds` will not attempt to add bonds between atoms that belong to different residues.
This is because in cases of suboptimal conformations or in very large structures atoms that are close in space may not be connected in the structure.
To override this behaviour, set the `restrict_residues` parameter to `False`.
.. code-block:: python
# add all bonds that are closer than 1.6 Å, even if they belong to different residues
glc.infer_bonds(bond_length=1.6, restrict_residues=False)
Residue Connections
-------------------
Alternatively, instead of `infer_bonds` one may use `infer_residue_connections` to get bonds that connect different residues.
This method will infer all bonds between atoms from different residues based on the distance between them. The inferred bonds are
saved in the `Molecule` and also returned in a `list`. If the optional argument `triplet` is set to `True`, the methodd will also
return the bonds immediately adjacent to the inferred bonds.
Take the following example of a molecule with two residues `A` and `B` that are connected by a bond between `OA` and `(1)CB`:
.. code-block:: text
connection --> OA OB --- H
/ \\ /
(1)CA --- (2)CA (1)CB
/ \\ \\
(6)CA (3)CA (2)CB --- (3)CB
\\ /
(5)CA --- (4)CA
If `triplet=False` the method will only return the bond between `OA` and `(1)CB`. However, if `triplet=True` it will also return the bond
between `(2)CA` and `OA` - thus forming a triplet of atoms `(2)CA`, `OA` and `(1)CB` that connect the two residues A and B.
.. code-block:: python
# infer all bonds between atoms from different residues
>>> glc.infer_residue_connections(triplet=False)
[(OA, (1)CB)]
>>> glc.infer_residue_connections(triplet=True)
[(OA, (1)CB), ((2)CA, OA)]
Adding residues and atoms
-------------------------
To add one or more new residue(s), we use the `add_residues` method, which expects a number `biobuild.Residue` objects as unnamed arguments.
Similarly, to add one or more new atom(s), we use the `add_atoms` method, which expects a number of `biobuild.Atom` objects as unnamed arguments.
Both methods allow to specify the parent object (`chain` or `residue`) via an optional argument and will automatically choose the last
chain or residue if none is specified.
.. code-block:: python
from biobuild import Residue, Atom
new_residue = Residue("XYZ", 1, " ")
# do things with the residue here
# ...
# add the residue to the molecule
# (add it to the last chain, whichever that may be)
glc.add_residues(new_residue)
new_atom = Atom("X", [0, 0, 0])
# add the atom to the first residue in the `glc` molecule
glc.add_atoms(new_atom, residue=1)
Removing residues and atoms
---------------------------
In order to remove residues or atoms or bonds, we can use the `remove_residues`, `remove_atoms` and `remove_bond`(yes, singluar!) methods.
They work exactly like their `add_` counterparts, but instead of adding, they remove the specified objects.
.. code-block:: python
# remove the first residue
glc.remove_residues(1)
# remove the first atom
glc.remove_atoms(1)
# remove the bond between the first and second atom
glc.remove_bond(1, 2)
.. warning::
When adding and removing atoms and residues, the `Molecule` object will not automatically update the `biopython.Structure` object as well as its internal
connectivity graph (the `AtomGraph`). However, if the user chooses to edit the biopython structure directly, the `AtomGraph` will **not** be updated automatically!
In this case, the user must call the `update_atom_graph` method to update the `AtomGraph` manually.
.. code-block:: python
# add a new residue to the molecule
glc.add_residues(new_residue)
# now add atoms into the residue directly instead of via `add_atoms`
new_residue.add(new_atom)
# now update the graph
glc.update_atom_graph()
Adjusting labelling
-------------------
Single-residue molecules that were loaded from a PDB file may not use the same atom labelling as the `PDBE` and `CHARMM` databases.
In order to quickly adjust the labelling, a method `autolabel` exists. `autolabel` uses the atom connectivity and a rule-based algorithm to infer
the most likely atom labels. However, since this method is not perfect, it is recommended to check the labelling afterwards and adjust it manually if necessary.
If working with large molecules that follow another labelling scheme it may be more efficient to simply
define your own linkage recipies or patches (see the documentation of `linkages`) that use the the appropriate labelling scheme.
.. code-block:: python
# load a molecule from a PDB file
glc = Molecule.from_pdb("glucose.pdb")
# adjust the labelling
glc.autolabel()
# save the molecule to a new PDB file
glc.to_pdb("glucose_autolabelled.pdb")
Another common operation is the adjustment of chain names and residue seqids. This can be done using the `reindex` method.
This method accepts three starting values for the chain name, residue seqid and atom serial number and will then reindex all chains, residues and atoms
to ensure they are continuously numbered and labelled. Some internal methods used when connecting different molecules are reliant on a continuous
numbering scheme, so this method should be called before connecting molecules that were loaded from PDB files.
.. code-block:: python
# load a molecule from a PDB file
glc = Molecule.from_pdb("glucose.pdb")
# reindex the molecule
glc.reindex()
We can also use one molecule as a "reference" for reindexing another molecule to make sure there are now labelling conflicts between them in case we want to connect them together later (this is usually done internally by _biobuild_ automatically).
.. code-block:: python
# load a molecule from a PDB file
glc = Molecule.from_pdb("glucose.pdb")
# load another molecule from a PDB file
cel = Molecule.from_pdb("cellulose.pdb")
cel.reindex() # make sure we have a continuous numbering scheme
# reindex the glucose molecule using the cellulose molecule as a reference
cel.adjust_indexing(glc)
Connecting Molecules
====================
Since most modifications are not simply single residues but rather complex structures, the second main purpose of a `Molecule` is to be easily
connected to other Molecules to form a larger structure. To this end, the `Molecule` class provides a number of methods to easily assemble complex structures from small single residue molecules.
Forming Polymers
----------------
The simplest way to generate a large structure is probably the `repeat` method, which will repeat the given molecule `n` times
to form a homo-polymer.
.. code-block:: python
# create a glucose molecule
glc = Molecule.from_compound("GLC")
# create cellulose from glucose
# using a 1-4 beta-beta glycosidic linkage
# which is pre-defined in the default CHARMMTopology
glc.repeat(10, "14bb")
# Now we have a cellulose of 10 glucoses
In the above example we used the `repeat` method explicitly, but we could also achieve the same with the short-hand `*=`. For this to work, we need to specify
the linkage type beforehand. We do this by setting the `patch` attribute before using any operator.
.. code-block:: python
# specify the "default" linkage type for connecting
# other molecules to this glucose
glc.linkage = "14bb"
# now make a cellulose by multiplying glucoses
glc *= 20
# Now we have a cellulose of 20 glucoses
If we wish to keep `glc` as a single residue Glucose and still get our desired cellulose,
we can set `inplace=False` when calling `repeat` or simply use the `*` operator, both of which
will have the same effect of creating a new copy that houses the appropriate residues.
.. code-block:: python
cel = glc.repeat(10, "14bb", inplace=False)
# or (equivalently)
glc.patch = "14bb"
cel = glc * 10
Connecting different Molecules
------------------------------
What if we want to connect different molecules? For example, we may want to connect a Galactose to a Glucose to form Lactose.
This can be achieved using the `attach` method, which will attach a given molecule to to another molecule.
.. code-block:: python
glc = Molecule.from_compound("GLC")
gal = Molecule.from_compound("GAL")
# attach the galactose to the glucose
# (we want a copy, so we set inplace=False just like with 'repeat')
lac = glc.attach(gal, "14bb", inplace=False)
# Now we have a lactose molecule
In the above example, the `attach` method is used to attach the galactose molecule to the glucose, but for those
among us who prefer a more shorty syntax, the `+` operator will do the same thing.
.. code-block:: python
# specify that incoming molecules shall be
# attached using a 1-4 beta linkage
glc.linkage = "14bb"
# now attach the galactose to the glucose
lac = glc + gal
Of course, if there is a `+` operator there should also be a `+=` operator, which is simply the equivalent of `attach` with `inplace=True`.
.. code-block:: python
glc.linkage = "14bb"
glc += gal
# Now 'glc' is a lactose molecule
Setting default Modifiers
-------------------------
So far, we have always worked with a 1-4 beta-beta glycosidic linkage, which we apparently could select using the string `"14bb"`.
But what if we want to use a different linkage type? For example, a 1-4 alpha-beta glycosidic linkage? You of course noticed,
that `attach` and `repeat` accept an argument `link` which allows you to specify the linkage type, and that if you leave it blank
the default linkage type is used. But how do we set the default linkage type?
Let's first check what linkage types are available by default anyway. Have you noticed an argument named `_topology`
at the end of the `attach` or `repeat` methods? The `topology` refers to the underlying _CHARMM_ topology which hosts the linkage type information.
By default a topology is already loaded in _biobuild_'s framework so it is not necessary for the user to specify anything here, but we can check which linkage types are available by:
.. code-block:: python
import biobuild as bb
# get the default topology
topology = bb.get_default_topology()
print(topology.patches)
Any of these linkages can be referenced by their name, e.g. `"14bb"` or `"14ab"`.
Wait a second, my desired linkage is not in the list! What now?! Well, you can always define a new `Linkage` to suit your needs.
Check out the documentation on :ref:`linkages` for more information on how to do this. If you have your desired linkage
ready to go, set it as the default by:
.. code-block:: python
my_molecule.linkage = my_linkage
# or if you feel "super correct"
my_molecule.set_linkage(my_linkage)
# or if you feel "extra cocky"
my_molecule % my_linkage # <- the modulo operator assignes the "modifier" to the molecule
Now any call to `attach`, `repeat`, or any of its operator proxies will use your defined linkage by default.
Setting the default Residue for attachment
------------------------------------------
When defining a `Linkage` we specify which atoms are supposed to be connected and removed, but we do not specify which residues
these atoms belong to. We specify this as arguments inside the `attach` method for instance, but we can also leave this blank, in which case the
last residue in the molecule is used by default. This is obviously not always what we want, however! Hence, if we do not want to specify the residue for attachment
at every `attach` call or if we want to use the `+` operator, we can set the default residue for attachment by setting the `attach_residue` attribute:
.. code-block:: python
# set the default attachment residue to the first residue
my_molecule.attach_residue = 1
# or
my_molecule.set_attach_residue(1)
# or (if you feel "extra cocky")
my_molecule @ 1 # <- the 'at' operator sets the attachment residue
# serial number indexing also works in reverse
# (set the second last residue as the default attachment residue)
my_molecule.attach_residue = -2
"""
from copy import deepcopy
import os
from typing import Union
import numpy as np
import Bio.PDB as bio
import biobuild.core.entity as entity
import biobuild.core.Linkage as Linkage
import biobuild.utils as utils
import biobuild.structural as structural
import biobuild.resources as resources
import biobuild.optimizers as optimizers
__all__ = [
"Molecule",
"read_pdb",
"write_pdb",
"read_cif",
"write_cif",
"read_smiles",
"molecule",
"connect",
"polymerize",
"phosphorylate",
"make_smiles",
"query_pubchem",
]
[docs]
def read_pdb(filename: str, id: str = None) -> "Molecule":
"""
Read a PDB file and return a molecule.
Parameters
----------
filename : str
The path to the PDB file
id : str
The id of the molecule
Returns
-------
molecule : Molecule
The molecule
"""
return Molecule.from_pdb(filename, id=id)
[docs]
def write_pdb(mol: "Molecule", filename: str) -> None:
"""
Write a molecule to a PDB file.
Parameters
----------
mol : Molecule
The molecule to write
filename : str
The path to the PDB file
"""
mol.to_pdb(filename)
[docs]
def read_cif(filename: str, id: str = None) -> "Molecule":
"""
Read a CIF file and return a molecule.
Parameters
----------
filename : str
The path to the CIF file
id : str
The id of the molecule
Returns
-------
molecule : Molecule
The molecule
"""
return Molecule.from_cif(filename, id=id)
[docs]
def write_cif(mol: "Molecule", filename: str) -> None:
"""
Write a molecule to a CIF file.
Parameters
----------
mol : Molecule
The molecule to write
filename : str
The path to the CIF file
"""
mol.to_cif(filename)
[docs]
def read_smiles(smiles: str, id: str = None) -> "Molecule":
"""
Read a SMILES string and return a molecule.
Parameters
----------
smiles : str
The SMILES string
Returns
-------
molecule : Molecule
The molecule
"""
return Molecule.from_smiles(smiles, id=id)
[docs]
def make_smiles(
mol: "Molecule", isomeric: bool = True, write_hydrogens: bool = False
) -> str:
"""
Generate a SMILES string from a molecule.
Parameters
----------
mol : Molecule
The molecule
isomeric : bool
Whether to include stereochemistry information
write_hydrogens : bool
Whether to include hydrogens in the SMILES string
Returns
-------
smiles : str
The SMILES string
"""
return mol.to_smiles(isomeric, write_hydrogens)
[docs]
def query_pubchem(query: str, by: str = "name") -> "Molecule":
"""
Query the PubChem database for a given
query string to obtain a Molecule object.
Parameters
----------
query : str
The query string
by : str
The type of query to perform. Can be one of:
he method to search by. This can be any of the following:
- cid
- name
- smiles
- sdf
- inchi
- inchikey
- formula
Returns
-------
Molecule or None
The molecule or None if no match was found
"""
try:
return Molecule.from_pubchem(query, by=by)
except:
return None
[docs]
def molecule(mol=None) -> "Molecule":
"""
Generate a molecule from an input. If the input is a string, the string can be a PDB id, some filename, SMILES or InChI string, IUPAC name or abbreviation.
This function will try its best to automatically generate the molecule with minimal user effort. However, using a dedicated classmethod is
recommended for more efficient and predictable results.
Parameters
----------
mol : str or structure-like object
An input string or structure-like object such as a BioPython Structure or RDKit Molecule, etc.
If nothing is provided, a new empty molecule is generated.
Returns
-------
molecule : Molecule
The generated molecule
Examples
--------
>>> from biobuild import molecule
>>> mol = molecule("GLC")
>>> mol = molecule("GLC.pdb")
>>> mol = molecule("alpha-d-glucose")
"""
if mol is None:
return Molecule.empty()
if isinstance(mol, bio.Structure.Structure) or isinstance(
mol, entity.base_classes.Structure
):
return Molecule(mol)
elif isinstance(mol, Molecule):
return mol.copy()
elif "openbabel" in str(type(mol).__mro__[0]):
return Molecule.from_pybel(mol)
elif "rdkit" in str(type(mol).__mro__[0]):
return Molecule.from_rdkit(mol)
elif (
"openmm" in str(type(mol).__mro__[0])
and hasattr(mol, "topology")
and hasattr(mol, "positions")
):
return Molecule.from_openmm(mol.topology, mol.positions)
if not isinstance(mol, str):
raise ValueError("input must be a structure object or a string")
# ------------------
# mol may be a PDB id
# ------------------
if resources.has_compound(mol):
return resources.get_compound(mol)
if os.path.isfile(mol):
_mol = mol.lower()
if _mol.endswith(".pdb"):
return Molecule.from_pdb(mol)
elif _mol.endswith(".cif"):
return Molecule.from_cif(mol)
elif _mol.endswith(".pkl"):
return Molecule.load(mol)
elif _mol.endswith(".json"):
return Molecule.from_json(mol)
elif (
_mol.endswith(".mol")
or _mol.endswith(".mol2")
or _mol.endswith(".sdf")
or _mol.endswith(".sd")
):
return Molecule.from_molfile(mol)
if " " not in mol:
try:
return Molecule.from_smiles(mol)
except:
pass
try:
return Molecule.from_pubchem(mol)
except:
pass
raise ValueError(f"Could not generate molecule from input: {mol}")
[docs]
def polymerize(
mol: "Molecule",
n: int,
link: Union[str, "Linkage.Linkage"] = None,
inplace: bool = False,
) -> "Molecule":
"""
Polymerize a molecule
Parameters
----------
mol : Molecule
The molecule to polymerize
n : int
The number of monomers to add
link : str or Linkage
The linkage to use for polymerization. If None, the default linkage of the molecule is used.
inplace : bool
Whether to polymerize the molecule in place or return a new molecule
Returns
-------
Molecule
The polymerized molecule
"""
if link is None and mol._linkage is None:
raise ValueError(
"No patch or recipe provided and no default is set on the molecule"
)
return mol.repeat(n, link, inplace=inplace)
[docs]
def connect(
mol_a: "Molecule",
mol_b: "Molecule",
link: Union[str, "Linkage.Linkage"],
at_residue_a: Union[int, "bio.Residue.Residue"] = None,
at_residue_b: Union[int, "bio.Residue.Residue"] = None,
copy_a: bool = True,
copy_b: bool = True,
_topology=None,
use_patch: bool = True,
) -> "Molecule":
"""
Connect two molecules together
Parameters
----------
mol_a : Molecule
The first (target) molecule
mol_b : Molecule
The second (source) molecule
link : Linkage or str
The linkage to use for connection. This can be either an instance of the Linkage class or a string identifier
of a pre-defined patch in the (currently loaded default or specified) CHARMMTopology.
at_residue_a : int or bio.PDB.Residue
The residue of the first molecule to connect to. If an integer is provided, the seqid must be used, starting at 1.
at_residue_b : int or bio.PDB.Residue
The residue of the second molecule to connect to. If an integer is provided, the seqid must be used, starting at 1.
copy_a : bool
Whether to copy the first molecule before connecting
copy_b : bool
Whether to copy the second molecule before connecting.
If False, all atoms of the second molecule will be added to the first molecule.
_topology : CHARMMTopology
A specific topology to use in case a pre-existing patch is used as link and only the string identifier
is supplied.
use_patch : bool
If the linkage has internal coordinates available (i.e. is a "patch") these are used by default. Set this to False
to force-use stitching and its associated conformational optimization instead.
Returns
-------
Molecule
The connected molecule
"""
if copy_b:
mol_b = mol_b.copy()
new = mol_a.attach(
mol_b,
link,
at_residue=at_residue_a,
other_residue=at_residue_b,
inplace=not copy_a,
_topology=_topology,
use_patch=use_patch,
)
return new
[docs]
def phosphorylate(
mol: "Molecule",
at_atom: Union[int, str, entity.base_classes.Atom],
delete: Union[int, str, entity.base_classes.Atom] = None,
inplace: bool = True,
) -> "Molecule":
"""
Phosphorylate a molecule at a specific atom
Parameters
----------
mol : Molecule
The molecule to phosphorylate
at_atom : int or str or Atom
The atom to phosphorylate. If an integer is provided, the atom seqid must be used, starting at 1.
delete : int or str or Atom
The atom to delete. If an integer is provided, the atom seqid must be used, starting at 1.
If not provided, any Hydrogen atom attached to the phosphorylated atom will be deleted.
inplace : bool
Whether to phosphorylate the molecule in place or return a new molecule
Returns
-------
Molecule
The phosphorylated molecule
"""
phos = Molecule.from_compound("PO4")
at_atom = mol.get_atom(at_atom)
delete = mol.get_atom(delete)
at_residue = at_atom.get_parent()
l = Linkage.Linkage("PHOS", "a custom phosphorylation")
l.add_bond(at_atom.id, "P")
if delete:
l.add_delete(delete.id, "target")
l.add_delete("O2", "source")
if not inplace:
mol = mol.copy()
mol.attach(phos, l, at_residue=at_residue)
return mol
def methylate(
mol: "Molecule",
at_atom: Union[int, str, entity.base_classes.Atom],
delete: Union[int, str, entity.base_classes.Atom] = None,
inplace: bool = True,
) -> "Molecule":
"""
Methylate a molecule at a specific atom
Parameters
----------
mol : Molecule
The molecule to methylate
at_atom : int or str or Atom
The atom to methylate. If an integer is provided, the atom seqid must be used, starting at 1.
delete : int or str or Atom
The atom to delete. If an integer is provided, the atom seqid must be used, starting at 1.
This atom needs to be in the same residue as the atom to methylate.
If not provided, any Hydrogen atom attached to the methylated atom will be deleted.
inplace : bool
Whether to methylate the molecule in place or return a new molecule
"""
methyl = Molecule.from_compound("CH3")
at_atom = mol.get_atom(at_atom)
at_residue = at_atom.get_parent()
delete = mol.get_atom(delete, residue=at_residue)
l = Linkage.Linkage("METHYLATE")
l.add_bond(at_atom.id, "C")
l.add_delete("HC1", "source")
if delete:
l.add_delete(delete.id, "target")
mol.attach(methyl, l, at_residue=at_residue, inplace=inplace)
return mol
[docs]
class Molecule(entity.BaseEntity):
"""
A molecule to add onto a scaffold.
A molecule consists of a single chain.
Parameters
----------
structure : bio.PDB.Structure
A biopython structure object
root_atom : str or int or Atom
The id or the serial number of the root atom
at which the molecule would be attached to a another
structure such as protein scaffold or another Molecule.
model : int
The model to use from the structure. Defaults to 0. This may be any
valid identifier for a model in the structure, such as an integer or string.
chain : str
The chain to use from the structure. Defaults to the first chain in the structure.
"""
def __init__(
self,
structure,
root_atom: Union[str, int, entity.base_classes.Atom] = None,
model: int = 0,
chain: str = None,
):
super().__init__(structure, model)
if not chain or len(self._model.child_list) == 1:
self._working_chain = self._model.child_list[0]
else:
self._working_chain = self._model.child_dict.get(chain)
if not self._working_chain:
raise ValueError("The chain {} is not in the structure".format(chain))
if root_atom:
self.set_root(root_atom)
if not root_atom in self._working_chain.get_atoms():
raise ValueError("The root atom is not in the structure")
self._root_atom = root_atom
[docs]
@classmethod
def empty(cls, id: str = None) -> "Molecule":
"""
Create an empty Molecule object
Parameters
----------
id : str
The id of the Molecule. By default an id is inferred from the filename.
Returns
-------
Molecule
An empty Molecule object
"""
structure = structural.make_empty_structure(id)
return cls(structure)
[docs]
@classmethod
def from_pdb(
cls,
filename: str,
root_atom: Union[str, int] = None,
id: str = None,
model: int = 0,
chain: str = None,
) -> "Molecule":
"""
Read a Molecule from a PDB file
Parameters
----------
filename : str
Path to the PDB file
root_atom : str or int
The id or the serial number of the root atom (optional)
id : str
The id of the Molecule. By default an id is inferred from the filename.
model : int
The model to use from the structure. Defaults to 0. This may be any
valid identifier for a model in the structure, such as an integer or string.
chain : str
The chain to use from the structure. Defaults to the first chain in the structure.
Returns
-------
Molecule
The Molecule object
"""
if id is None:
id = utils.filename_to_id(filename)
struct = utils.defaults.__bioPDBParser__.get_structure(id, filename)
new = cls(struct, root_atom, model=model, chain=chain)
bonds = utils.pdb.parse_connect_lines(filename)
if len(bonds) != 0:
for b in bonds:
new.add_bond(*b)
return new
[docs]
@classmethod
def from_compound(
cls,
compound: str,
by: str = "id",
root_atom: Union[str, int] = None,
) -> "Molecule":
"""
Create a Molecule from a reference compound from the PDBECompounds database
Parameters
----------
compound : str
The compound to search for
by : str
The field to search by. This can be
- "id" for the PDB id
- "name" for the name of the compound (must match any known synonym of the iupac name)
- "formula" for the chemical formula
- "smiles" for the SMILES string (also accepts InChI)
root_atom : str or int
The id or the serial number of the root atom (optional)
"""
mol = resources.get_default_compounds().get(compound, by=by)
if isinstance(mol, list):
raise ValueError(
f"Multiple compounds found using '{by}={compound}', choose any of these ids specifically {[i.id for i in mol]}"
)
mol.set_root(root_atom)
return mol
[docs]
@classmethod
def from_smiles(
cls,
smiles: str,
id: str = None,
root_atom: Union[str, int] = None,
add_hydrogens: bool = True,
) -> "Molecule":
"""
Read a Molecule from a SMILES string
Parameters
----------
smiles : str
The SMILES string
id : str
The id of the Molecule. By default the provided smiles string is used.
root_atom : str or int
The id or the serial number of the root atom (optional)
add_hydrogens : bool
Whether to add hydrogens to the molecule
Returns
-------
Molecule
The Molecule object
"""
obj = structural.read_smiles(smiles, add_hydrogens)
if structural.smiles.use_rdkit:
new = cls.from_rdkit(obj, id=id)
elif structural.smiles.use_openbabel:
obj.title = id
new = cls.from_pybel(obj)
new.set_root(root_atom)
return new
[docs]
@classmethod
def from_pubchem(
cls,
query: str,
root_atom: Union[str, int] = None,
by: str = "name",
idx: int = 0,
) -> "Molecule":
"""
Create a Molecule from PubChem
Note
----
PubChem follows a different atom labelling scheme than the CHARMM force field!
This means that atom names may not match the names required by the default patches that are
integrated in biobuild. It is advisable to run `autolabel` or `relabel_hydrogens` on the molecule.
Naturally, custom patches or recipies working with adjusted atom names will always work.
Parameters
----------
query : str
The query to search for in the PubChem database
root_atom : str or int
The id or the serial number of the root atom (optional)
by : str
The method to search by. This can be any of the following:
- cid
- name
- smiles
- sdf
- inchi
- inchikey
- formula
idx : int
The index of the result to use if multiple are found. By default, the first result is used.
Returns
-------
Molecule
The Molecule object
"""
_compound_2d, _compound_3d = resources.pubchem.query(query, by=by, idx=idx)
new = _molecule_from_pubchem(_compound_2d.iupac_name, _compound_3d)
_new = cls(new.structure)
_new.add_bonds(*(i.to_tuple() for i in new._bonds))
new = _new
new.id = _compound_2d.iupac_name
new.set_root(root_atom)
return new
[docs]
def to_smiles(self, isomeric: bool = True, write_hydrogens: bool = False) -> str:
"""
Convert the molecule to a SMILES string
Parameters
----------
isomeric : bool
Whether to include stereochemistry information in the SMILES string
write_hydrogens : bool
Whether to include hydrogens in the SMILES string
Returns
-------
str
The SMILES string
"""
return structural.make_smiles(
self, isomeric=isomeric, add_hydrogens=write_hydrogens
)
[docs]
def get_residue_connections(
self,
residue_a=None,
residue_b=None,
triplet: bool = True,
direct_by: str = None,
):
"""
Get bonds between atoms that connect different residues in the structure
This method is different from `infer_residue_connections` in that it works
with the already present bonds in the molecule instead of computing new ones.
Parameters
----------
residue_a, residue_b : Union[int, str, tuple, bio.Residue.Residue]
The residues to consider. If None, all residues are considered.
Otherwise, only between the specified residues are considered.
triplet : bool
Whether to include bonds between atoms that are in the same residue
but neighboring a bond that connects different residues. This is useful
for residues that have a side chain that is connected to the main chain.
This is mostly useful if you intend to use the returned list for some purpose,
because the additionally returned bonds are already present in the structure
from inference or standard-bond applying and therefore do not actually add any
particular information to the Molecule object itself.
direct_by : str
The attribute to sort by. Can be either "serial", "resid" or "root".
In the case of "serial", the bonds are sorted by the serial number of the first atom.
In the case of "resid", the bonds are sorted by the residue id of the first atom.
In the case of "root", the bonds are sorted by the root atom of the first atom.
Set to None to not sort the bonds.
Returns
-------
set
A set of tuples of atom pairs that are bonded and connect different residues
"""
bonds = super().get_residue_connections(
residue_a=residue_a, residue_b=residue_b, triplet=triplet
)
if direct_by is not None:
direct_connections = None
if direct_by == "resid":
direct_connections = super().get_residue_connections(
residue_a=residue_a, residue_b=residue_b, triplet=False
)
direct_connections1, direct_connections2 = set(
a for a, b in direct_connections
), set(b for a, b in direct_connections)
direct_connections = direct_connections1.union(direct_connections2)
bonds = self._direct_bonds(bonds, direct_by, direct_connections)
return bonds
[docs]
def repeat(self, n: int, link=None, inplace: bool = True):
"""
Repeat the molecule n times into a homo-polymer.
Parameters
----------
n : int
The number or units of the final polymer.
link : str or Patch or Recipe
The patch or recipe to use when patching individual units together.
If noe is given, the default patch or recipe is used (if defined).
inplace : bool
If True the molecule is directly modified, otherwise a copy of the molecule is returned.
Returns
-------
molecule
The modified molecule (either the original object or a copy)
"""
if not isinstance(n, int):
raise TypeError("Can only multiply a molecule by an integer")
if n <= 0:
raise ValueError("Can only multiply a molecule by a positive integer")
if not link and not self._linkage:
raise RuntimeError("Cannot multiply a molecule without a patch defined")
_other = self.copy()
if not inplace:
obj = self.copy()
else:
obj = self
_patch = obj._linkage
if link:
obj % link
for i in range(n - 1):
obj += _other
if _patch:
obj % _patch
return obj
[docs]
def attach(
self,
other: "Molecule",
link: Union[str, "Linkage.Linkage"] = None,
at_residue: Union[int, "entity.base_classes.Residue"] = None,
other_residue: Union[int, "entity.base_classes.Residue"] = None,
use_patch: bool = True,
inplace: bool = True,
other_inplace: bool = False,
_topology=None,
):
"""
Attach another structure to this one using a Patch or a Recipe.
Parameters
----------
other : Molecule
The other molecule to attach to this one
link : str or Linkage
Either a Patch to apply when attaching or a Recipe to use when stitching.
If None is defined, the default patch or recipe that was set earlier on the molecule is used.
at_residue : int or Residue
The residue to attach the other molecule to. If None, the defined `attach_residue` is used.
other_residue : int or Residue
The residue in the other molecule to attach this molecule to. If None, the defined `attach_residue` of the other molecule is used.
use_patch : bool
If the specified linkage is a patch (has internal coordinates) it can and is by default applied as a patch. However, it can also be used as a recipe.
Set this to false if you want to use the patch as a recipe.
inplace : bool
If True the molecule is directly modified, otherwise a copy of the molecule is returned.
other_inplace : bool
All atoms from the other molecule are integrated into this one. Hence, the other molecule is left empty. If False, a copy of the other molecule is used.
Thus leaving the original molecule intact.
_topology : Topology
The topology to use when attaching. If None, the topology of the molecule is used. Only used if the patch is a string.
"""
if not isinstance(other, Molecule):
raise TypeError("Can only attach a Molecule to another Molecule")
if not inplace:
obj = self.copy()
else:
obj = self
if not link:
link = obj._linkage
if not link:
raise ValueError("Cannot attach a molecule without a patch defined")
if not other_inplace:
_other = other.copy()
else:
_other = other
if isinstance(link, str):
if not _topology:
_topology = resources.get_default_topology()
link = _topology.get_patch(link)
if link.has_IC and use_patch:
obj.patch_attach(
_other,
link,
at_residue=at_residue,
other_residue=other_residue,
_topology=_topology,
)
else:
obj.stitch_attach(
_other,
link,
at_residue=at_residue,
other_residue=other_residue,
)
return obj
[docs]
def patch_attach(
self,
other: "Molecule",
patch: Union["Linkage.Linkage", str] = None,
at_residue: Union[int, bio.Residue.Residue] = None,
other_residue: Union[int, bio.Residue.Residue] = None,
_topology=None,
):
"""
Attach another structure to this one using a Patch.
Parameters
----------
other : Molecule
The other molecule to attach to this one
patch : str or Linkage
A linkage to apply when attaching. If none is given, the default link that was set earlier
on the molecule is used. If no patch was set, an AttributeError is raised. If a string is
given, it is interpreted as the name of a patch in the topology.
at_residue : int or Residue
The residue to attach the other molecule to. If None, the last residue of the molecule.
other_residue : int or Residue
The residue of the other molecule to attach. If None, the first residue of the other molecule.
_topology
A specific topology to use for referencing.
If None, the default CHARMM topology is used.
"""
if not patch:
patch = self._linkage
if not patch:
raise AttributeError(
"No patch was set for this molecule. Either set a default patch or provide a patch when attaching."
)
if isinstance(patch, str):
if not _topology:
_topology = resources.get_default_topology()
patch = _topology.get_patch(patch)
if not patch:
raise ValueError(
"No patch was found with the given name. Either set a default patch or provide a patch when attaching."
)
p = structural.__default_keep_keep_patcher__
p.apply(patch, self, other, at_residue, other_residue)
p.merge()
return self
[docs]
def stitch_attach(
self,
other: "Molecule",
recipe: "Linkage.Linkage" = None,
remove_atoms=None,
other_remove_atoms=None,
at_atom=None,
other_at_atom=None,
at_residue=None,
other_residue=None,
):
"""
Stitch two molecules together by removing atoms and connecting them with a bond. This works without a pre-defined patch.
Parameters
----------
other : Molecule
The other molecule to attach to this one
recipe : Recipe
The recipe to use when stitching. If None, the default recipe that was set earlier on the molecule is used (if defined).
remove_atoms : list of int
The atoms to remove from this molecule. Only used if no recipe is provided.
other_remove_atoms : list of int
The atoms to remove from the other molecule. Only used if no recipe is provided.
at_atom : int or str or Bio.PDB.Atom
The atom forming the bond in this molecule. If a string is provided, an `at_residue` needs to be defined from which to get the atom. If None is provided, the root atom is used (if defined). Only used if no recipe is provided.
other_at_atom : int or str or Bio.PDB.Atom
The atom to attach to in the other molecule. If a string is provided, an `other_residue` needs to be defined from which to get the atom. If None is provided, the root atom is used (if defined). Only used if no recipe is provided.
at_residue : int or Residue
The residue to attach the other molecule to. If None, the `attach_residue` is used. Only used if a recipe is provided and the atoms
"""
if not recipe and remove_atoms is None:
if not self._linkage:
raise AttributeError(
"No recipe was set for this molecule and no manual instructions were found. Either set a default recipe, provide a recipe when stitching, or provide the information about removed and bonded atoms directly."
)
recipe = self._linkage
if recipe:
target_atom, source_atom = recipe._stitch_ref_atoms
return self.stitch_attach(
other,
remove_atoms=recipe.deletes[0],
other_remove_atoms=recipe.deletes[1],
at_atom=target_atom,
other_at_atom=source_atom,
at_residue=at_residue,
other_residue=other_residue,
)
if not isinstance(other, Molecule):
raise TypeError("Can only stitch two molecules together")
if not at_atom:
at_atom = self._root_atom
if not at_atom:
raise ValueError(
"No atom to attach to was provided and no root atom was defined in this molecule"
)
if not other_at_atom:
other_at_atom = other._root_atom
if not other_at_atom:
raise ValueError(
"No atom to attach to was provided and no root atom was defined in the other molecule"
)
if not at_residue:
at_residue = self.attach_residue
if not other_residue:
other_residue = other.attach_residue
# since we are already copying before we can use the keep-keep stitcher, actually...
p = structural.__default_keep_keep_stitcher__
p.apply(
self,
other,
remove_atoms,
other_remove_atoms,
at_atom,
other_at_atom,
at_residue,
other_residue,
)
self = p.merge()
return self
[docs]
def optimize(
self,
residue_graph: bool = None,
algorithm: str = None,
rotatron: str = None,
rotatron_kws: dict = None,
algorithm_kws: dict = None,
inplace: bool = True,
):
"""
Optimize the molecule's conformation. This is a convenience method with less
customizability than a manual optimization using the `optimizers` module.
Parameters
----------
residue_graph : bool
Whether to use the residue graph or the full atom graph for optimization.
The residue graph is faster but less accurate. If the molecule is larger than 100 atoms,
the residue graph is used by default.
algorithm : str
The optimization algorithm to use. If not provided, an algorithm is determined based on the molecule's size.
This can be one of the following:
- "genetic" for a genetic algorithm
- "scipy" for a scipy-implemented gradient-based optimization
- "swarm" for a particle swarm optimization
- "anneal" for a simulated annealing optimization
- "rdkit" for an RDKit-implemented force-field-based optimization (if RDKit is installed)
rotatron : str
The rotatron to use. This can be one of the following:
- "distance" for a distance-based rotatron (default)
- "overlap" for an overlap-based rotatron
algorithm_kws : dict
Keyword arguments to pass to the optimization algorithm
rotatron_kws : dict
Keyword arguments to pass to the rotatron
inplace : bool
Whether to optimize the molecule in place or return a copy.
Returns
-------
molecule
The optimized molecule (either the original object or a copy)
"""
if residue_graph is None:
residue_graph = self.count_atoms() > 100
if not algorithm_kws:
algorithm_kws = {}
if not rotatron_kws:
rotatron_kws = {}
algorithm = algorithm or optimizers.auto_algorithm(self)
if algorithm == "rdkit":
obj = self.copy() if not inplace else self
out = optimizers.rdkit_optimize(obj)
out.id = self.id
return out
elif algorithm == "scipy":
algorithm = optimizers.scipy_optimize
elif algorithm == "genetic":
algorithm = optimizers.genetic_optimize
elif algorithm == "swarm":
algorithm = optimizers.swarm_optimize
elif algorithm == "anneal":
algorithm = optimizers.anneal_optimize
else:
raise ValueError(f"Unknown algorithm {algorithm}")
rotatron = rotatron or "distance"
if rotatron == "distance":
rotatron = optimizers.DistanceRotatron
elif rotatron == "overlap":
rotatron = optimizers.OverlapRotatron
else:
raise ValueError(f"Unknown rotatron {rotatron}")
obj = self.copy() if not inplace else self
if residue_graph:
graph = obj.make_residue_graph(detailed=True)
else:
graph = obj.make_atom_graph(_copy=False)
edges = graph.find_rotatable_edges(
graph.central_node, min_ancestors=1, min_descendants=1
)
env = rotatron(graph, edges, **rotatron_kws)
sol, _ = algorithm(env, **algorithm_kws)
if sol.shape[0] != len(env.rotatable_edges):
sol = sol[0]
out = optimizers.apply_solution(sol, env, obj)
return out
def __add__(self, other) -> "Molecule":
"""
Add two molecules together. This will return a new molecule.
"""
if isinstance(other, (tuple, list)):
new = self + other[0]
for o in other[1:]:
new += o
return new
elif isinstance(other, Molecule):
patch = self._linkage
if not patch:
patch = other._linkage
if not patch:
raise RuntimeError(
"Cannot add two molecules together without a patch, set a default patch on either of them (preferably on the one you are adding to, i.e. mol1 in mol1 + mol2)"
)
new = self.attach(other, patch, inplace=False, other_inplace=False)
return new
new = self.copy()
if isinstance(other, entity.base_classes.Atom):
new.add_atoms(other)
elif isinstance(other, entity.base_classes.Residue):
new.add_residues(other)
elif isinstance(other, entity.base_classes.Chain):
new._model.add(other)
elif isinstance(other, entity.base_classes.Model):
new._base_struct.add(other)
return new
def __iadd__(self, other) -> "Molecule":
"""
Attach another molecule to this one
"""
if isinstance(other, (tuple, list)):
for o in other:
self += o
return self
elif isinstance(other, Molecule):
patch = self._linkage
if not patch:
patch = other._linkage
if not patch:
raise RuntimeError(
"Cannot add two molecules together without a patch, set a default patch on either of them (preferably on the one you are adding to, i.e. mol1 in mol1 += mol2)"
)
self.attach(other, patch)
return self
elif isinstance(other, entity.base_classes.Atom):
self.add_atoms(other)
elif isinstance(other, entity.base_classes.Residue):
self.add_residues(other)
elif isinstance(other, entity.base_classes.Chain):
self._model.add(other)
elif isinstance(other, entity.base_classes.Model):
self._base_struct.add(other)
return self
def __sub__(self, other) -> "Molecule":
"""
Remove atoms, residues, chains or models from the molecule
"""
if isinstance(other, (tuple, list)):
new = self.copy()
for o in other:
new -= o
return new
new = self.copy()
if isinstance(other, entity.base_classes.Atom):
new.remove_atoms(other)
elif isinstance(other, entity.base_classes.Residue):
new.remove_residues(other)
elif isinstance(other, entity.base_classes.Chain):
new._model.detach_child(other.get_id())
elif isinstance(other, entity.base_classes.Model):
new._base_struct.detach_child(other.get_id())
elif isinstance(other, Molecule):
new.remove_residues(*other.residues)
return new
def __isub__(self, other) -> "Molecule":
"""
Remove atoms, residues, chains or models from the molecule
"""
if isinstance(other, (tuple, list)):
for o in other:
self -= o
return self
elif isinstance(other, entity.base_classes.Atom):
self.remove_atoms(other)
elif isinstance(other, entity.base_classes.Residue):
self.remove_residues(other)
elif isinstance(other, entity.base_classes.Chain):
self._model.detach_child(other.get_id())
elif isinstance(other, entity.base_classes.Model):
self._base_struct.detach_child(other.get_id())
elif isinstance(other, Molecule):
self.remove_residues(*other.residues)
return self
def __mul__(self, n) -> "Molecule":
"""
Add multiple identical molecules together using the * operator (i.e. mol * 3)
This requires that the molecule has a patch defined
"""
if not self._linkage:
raise RuntimeError("Cannot multiply a molecule without a patch defined")
if not isinstance(n, int):
raise TypeError("Can only multiply a molecule by an integer")
if n <= 0:
raise ValueError("Can only multiply a molecule by a positive integer")
new = self.attach(self, self._linkage, inplace=False, other_inplace=False)
for i in range(n - 2):
new = new.attach(self, self._linkage, inplace=True, other_inplace=False)
return new
def __imul__(self, n) -> "Molecule":
"""
Add multiple identical molecules together using the *= operator (i.e. mol *= 3)
This requires that the molecule has a patch defined
"""
self.repeat(n)
return self
def __repr__(self):
return f"Molecule({self.id})"
def __str__(self):
string = f"Molecule {self.id}:\n"
if len(self.residues) > 1:
string += str(self.make_residue_graph())
else:
string += str(self.residues[0])
return string
def _molecule_from_pubchem(id, comp):
"""
Convert a PubChem compound to a Molecule.
Parameters
----------
id : str
The id of the molecule
comp : pcp.Compound
The PubChem compound
Returns
-------
Molecule
The molecule
"""
bonds = comp.bonds
mol = Molecule.empty(id)
residue = entity.base_classes.Residue("UNK", " ", 1)
mol.add_residues(residue)
# structure = bio.Structure.Structure(id)
# model = bio.Model.Model(0)
# structure.add(model)
# chain = bio.Chain.Chain("A")
# model.add(chain)
# residue = bio.Residue.Residue((" ", 1, " "), "UNK", 1)
# chain.add(residue)
element_counts = {}
adx = 1
for atom in comp.atoms:
element = atom.element
element_counts[element] = element_counts.get(element, 0) + 1
id = f"{element}{element_counts[element]}"
_atom = entity.base_classes.Atom(
id,
coord=np.array((atom.x, atom.y, atom.z)),
serial_number=adx,
bfactor=0.0,
occupancy=0.0,
element=element.upper(),
fullname=id,
altloc=" ",
pqr_charge=0.0,
)
residue.add(_atom)
adx += 1
for bond in bonds:
mol.add_bond(bond.aid1, bond.aid2, bond.order)
return mol
# def _molecule_from_ctfile(id, ctfile):
# """
# Convert a CTFile object to a Molecule.
# Parameters
# ----------
# id : str
# The id of the molecule
# ctfile : CTFile
# The CTFile object
# Returns
# -------
# Molecule
# The molecule
# """
# new = Molecule.empty(id)
# residue = entity.base_classes.Residue("UNK", " ", 1)
# new.add_residues(residue)
# element_counts = {}
# for atom in ctfile.atoms:
# element_counts.setdefault(atom.atom_symbol, 1)
# count = element_counts[atom.atom_symbol]
# element_counts[atom.atom_symbol] += 1
# x = float(atom._ctab_data["x"])
# y = float(atom._ctab_data["y"])
# z = float(atom._ctab_data["z"])
# _atom = entity.base_classes.Atom(
# id=f"{atom.atom_symbol}{count}",
# coord=np.array((x, y, z)),
# serial_number=int(atom.atom_number),
# pqr_charge=float(atom.charge),
# )
# residue.add(_atom)
# for bond in ctfile.bonds:
# a, b = bond.first_atom, bond.second_atom
# a = int(a.atom_number)
# b = int(b.atom_number)
# order = int(bond._ctab_data["bond_type"])
# for i in range(order):
# new.add_bond(a, b)
# return new
if __name__ == "__main__":
# ser = Molecule.from_json("ser.json")
# link = Linkage.from_json("peptide_linkage.json")
# pep = ser.attach(ser, link, use_patch=False)
# pep.show()
# # ser = Molecule.from_pubchem("SER")
# ser.to_cif("ser.cif")
# ser.to_pdb("ser.pdb")
# recipe = Linkage()
# recipe.add_delete("O1", "target")
# recipe.add_delete("HO1", "target")
# recipe.add_delete("HO4", "source")
# recipe.add_bond(("C1", "O4"))
# glc = Molecule.from_compound("GLC")
# glc.set_linkage(recipe)
# glc2 = glc.copy()
# _current_residues = len(glc.residues)
# glc3 = glc + glc2
# glc3.show()
# glc3 = glc % "14bb" + glc
pass
# import pickle
# tmp = pickle.load(open("/Users/noahhk/GIT/biobuild/tmp.pickle", "rb"))
# tmp.get_residue_connections()
# exit()
# from timeit import timeit
# # man = Molecule.from_compound("MAN")
# glc = Molecule.from_compound("GLC")
# # man.adjust_indexing(glc)
# # assert glc.atoms[0].serial_number == len(man.atoms) + 1
# t1 = timeit()
# glc.repeat(5, "14bb")
# # glc % "14bb"
# # glc *= 10
# t2 = timeit()
# print(t2 - t1)
from biobuild.utils import visual
# v = visual.MoleculeViewer3D(glc)
# v.show()
# man = Molecule.from_pdb("support/examples/membrane.pdb", model=4)
# man.infer_bonds()
# man.
man = Molecule.from_pubchem("benzene")
man.infer_bonds(restrict_residues=False)
g = man.make_residue_graph(True)
v = visual.MoleculeViewer3D(g)
for c in man.get_residue_connections():
v.draw_vector(
None, c[0].coord, 1.3 * (c[1].coord - c[0].coord), color="magenta"
)
v.show()