Setting carbon chirality upon ketoreduction of hydroxylated PKS chains is not possible

[BTheDragonMaster]

Currently, it is not possible to use KR domain subtypes that are not 'UNKNOWN' or 'C1' when the carbon adjacent to the side chain is hydroxylated or otherwise modified, as the function checks for the presence of a hydrogen neighbour of this carbon:

In pks_tailoring_reactions.py, ketoreduction function:

        for atom in chiral_c.neighbours:
            if atom != c_1 and atom != c_3 and atom.type != 'H' and \
                    not atom.annotations.in_central_chain:
                first_sidechain_atom = atom
            elif atom.type == 'H':
                hydrogen = atom

        assert hydrogen
        assert first_sidechain_atom

It has currently not led to any errors, as all trans-AT modules that apply tailoring followed by ketoreduction use the 'UNKNOWN' domain subtype, but it may lead to errors in the future.