Source code for buildamol.extensions.bio.glycans.glycan

from .iupac import IUPACParser
from buildamol import core, resources

resources.load_sugars()

__all__ = ["glycan"]




def glycan(iupac: str, id: str = "UNK") -> core.Molecule:
    """
    Create a glycan molecule from an IUPAC string

    Parameters
    ----------
    iupac : str
        The IUPAC string of the glycan
    id : str
        The id of the glycan molecule

    Returns
    -------
    Molecule
        The glycan molecule
    """
    parser = IUPACParser()
    segments = parser.parse(iupac)
    return _segments_to_mol(id, segments)



def _segments_to_mol(id, glycan_segments, _topology=None):
    """
    Make a molecule from a list of glycan segments that were generated by the IUPACParser class

    Parameters
    ----------
    id : str
        The id of the molecule
    glycan_segments : list
        A list of glycan segments
    Returns
    -------
    Molecule
        The molecule
    """
    if not _topology:
        _topology = resources.get_default_topology()

    # Check that all segments have a known patch
    for segment in glycan_segments:
        link = segment[-1]

        if not _topology.has_patch(link):
            # make a linkage
            _link = core.linkage(
                f"O{link[0]}",
                f"C{link[1]}",
                [f"HO{link[0]}"],
                [f"O{link[1]}", f"HO{link[1]}"],
                id=link,
            )
            _topology.add_patch(_link)

    mol = None
    first_mol = None
    second_mol = None
    at_residue = None
    other_residue = None
    residue_id_mapping = {}
    for i, segment in enumerate(glycan_segments):
        first, second, link = segment

        first_name = first.split("@")[0]
        second_name = second.split("@")[0]

        if first in residue_id_mapping:
            at_residue = residue_id_mapping[first]
            first_mol = mol
        else:
            if first_name.startswith("b-") and first_name[2:].isupper():
                first_name = first_name[2:]
            first_mol = core.molecule(first_name)
            if isinstance(first_mol, list):
                first_mol = first_mol[0]

            # if we did not get the compound from the PDBE compounds,
            # we probably got them from PubChem, in which case we need to autolabel them
            if not resources.has_compound(first_mol.id):
                first_mol.autolabel()
            residue_id_mapping[first] = len(residue_id_mapping) + 1
            at_residue = None

        if second in residue_id_mapping:
            other_residue = residue_id_mapping[second]
            second_mol = mol
        else:
            if second_name.startswith("b-") and second_name[2:].isupper():
                second_name = second_name[2:]
            second_mol = core.molecule(second_name)
            if isinstance(second_mol, list):
                second_mol = second_mol[0]

            if not resources.has_compound(second_mol.id):
                second_mol.autolabel()
            residue_id_mapping[second] = len(residue_id_mapping) + 1
            other_residue = None

        if not mol:
            mol = first_mol
        mol.attach(
            second_mol,
            link,
            at_residue=at_residue,
            other_residue=other_residue,
            _topology=_topology,
        )

    mol.id = id
    return mol


if __name__ == "__main__":

    iupac = "Gal(a1-4)Gal(b1-4)GlcNAc(b1-6)GalNAc(b1-"
    mol = glycan(iupac)
    mol.show()