pyrosetta_help.ligands package

Submodules

pyrosetta_help.ligands.hunter module

class pyrosetta_help.ligands.hunter.LigandHunter(sequence: str)

Bases: object

Given a sequence find homologues (Blast) and find if they have ligands (PDBe API). The definition of what ligand is a cofactor comes from PDBe and does not count ions or triphospho-nucleotides as cofactors, but does count NADH adducts.

• .data is a list of dictionaries.

• .to_dataframe() converts into a pandas dataframe for further analysis.

• .candidate_ligands list of ligand residue 3-letter codes

__init__(sequence: str)

Parameters:

sequence – is assumed clean protein sequence

property candidate_ligands: List[str]

property cofactor_codes

property cofactor_reference: Dict[str, List[dict]]

get_most_common_ligands() → List[Tuple[str, int]]

get_pdb_entry_by_ligand(ligand_code: str) → dict

get pdb entry by ligand. Returns the first, which should be lowest e-value

property ligand_data

Data for the ligands.

Note that LigandNicker has a get smiles method, that works like this, but is unrelated.

missing_codes = ['ATP', 'GTP', 'CA', 'MG', 'W']

to_dataframe()

Converts .data to a pandas dataframe

pyrosetta_help.ligands.load module

pyrosetta_help.ligands.load.get_smiles(ligand_code: str) → str

Get the smiles of a ligand. Remember that PDBe smiles need to charged to pH 7.

pyrosetta_help.ligands.load.parameterized_pose_from_file(pdb_filename, wanted_ligands: List[str] | Dict[str, str | None] = (), force_parameterisation: bool = False, neutralize_params: bool = True, save_params: bool = True, overriding_params=()) → Pose

pose loading, the circutous way to not lose ligand or use PDB_component. Assumes all mystery components are PDB residues. Works best with ignore_unrecognized_res False

Parameters:

• pdb_filename –

• wanted_ligands – a list of three-letter codes (PDB chemcomponent identifiers) or a dictionary of three-letter codes to None or smiles

• force_parameterisation –

• neutralize_params – protonated for pH 7

• save_params –

• overriding_params – list of params filenames

Returns:

pyrosetta_help.ligands.load.parameterized_pose_from_pdbblock(pdbblock: str, wanted_ligands: List[str] | Dict[str, str | None] = (), force_parameterisation: bool = False, neutralize_params: bool = True, save_params: bool = True, overriding_params=()) → Pose

pose loading, the circutous way to not loose ligand or use PDB_component. Assumes all mystery components are PDB residues. Works best with ignore_unrecognized_res False

Parameters:

• pdbblock –

• wanted_ligands – a list of three letter codes or a dictionary of three letter codes to None or smiles

• force_parameterisation –

• neutralize_params – protonated for pH 7

• save_params –

• overriding_params – list of params filenames

Returns:

pyrosetta_help.ligands.nick module

class pyrosetta_help.ligands.nick.LigandNicker(pdb_filename: None | str = None, pdb_filehandle: IOBase | None = None, pdb_block: None | str = None, pose: Pose | None = None, chain: str = 'A', wanted_ligands: List[str] = (), force_parameterisation: bool = False, neutralize_params: bool = True, save_params: bool = True, overriding_params=())

Bases: object

Given a pdb_file (regular initialisation) or code (.from_pdbcode classmethod) and a list of 3-letter codes of wanted residues (wanted_ligands argument), it loads it as Pyrosetta Pose (donor_pose), ready for migrate to loads the acceptor_pose and nick the residues that are wanted.

If force_parameterisation is on it or the residue is novel it parameterises it.

__init__(pdb_filename: None | str = None, pdb_filehandle: IOBase | None = None, pdb_block: None | str = None, pose: Pose | None = None, chain: str = 'A', wanted_ligands: List[str] = (), force_parameterisation: bool = False, neutralize_params: bool = True, save_params: bool = True, overriding_params=())

Initialisation loads the donor. migrate loads the acceptor.

Unfortunately, I originally wrote it to use pdb_filename only. I should have written the init to accept a pose and a class method the filename. And now some chucks of out there in the wild use pdb_filename. So I could not switch it to something generic. Now it accepts four possible alternative choices: pdb_filename, pdb_filehandle, pdb_block, pose.

Parameters:

• pdb_filename –

• pdb_filehandle –

• pdb_block –

• pose –

• chain –

• wanted_ligands –

• force_parameterisation –

• neutralize_params – pH 7 protonation

• save_params –

• overriding_params – overide paramaterisation and use provide params

constrain_migrated(wanted_vector, dex)

classmethod from_pdbcode(pdb_code: str, chain: str, *args, **kvargs)

get_mapping_between_poses(index_list: List[int]) → Dict[int, int]

Given a list of indices for one pose, return their aligned equivalents in the second. Modded to be donor_pose –> acceptor_pose

get_surrounding_residue(pose, chain_filter, wanted_selector)

get_wanted_selector()

make_atomID_map(dex, query_pose, target_pose)

make_constraint_foreign_hbond(hbond, dex: Dict[int, int]) → None | AtomPairConstraint

migrate(acceptor_pose: Pose, acceptor_chain: str = 'A', constrained: bool = True, relaxed: bool = True, relax_radius: int = 20, relax_cycles: int = 3)

The acceptor pose is the non-empty pose.

This method aligns the sequences of the acceptor and donor pose. It finds the mapping of the neighbourhood of the wanted residues of the donor_pose It superimposes the poses by those residues. It adds the residues that need nicking. It adds constraints (optionally) based on the hydrogen bonding of the residues around the wanted residues. onto the .acceptor_pose. To check:

>>> print( len(self.acceptor_pose.constraint_set().get_all_constraints())  )

It then optionally relaxes the neighbourhood.

Parameters:

• acceptor_pose –

• acceptor_chain –

• constrained –

• relaxed –

• relax_radius –

• relax_cycles –

Returns:

relax_migrated(distance: int = 20, cycles: int = 3, atom_pair_weight: int = 5)

pyrosetta_help.ligands.nick.chain_letter_to_number(letter, pose)

Module contents