In the original code, the padded_mol did not necessarily contain all hydrogens.

Although the PR works, it requires all hydrogens, so it does not reproduce REMARK INDEX MAP (mapidx_pad).

And therefore it also breaks test test_export_sidechains_no_idxmap that relies on old indices, although with new indices this PR should work.

It seems like the tests (test_export_sidechains_no_idxmap and test_AHHY_flex_residues) may rely on the exact REMARK INDEX MAP.

Closing this, while it technically works, it changes atom indexing and hydrogen count in a way that could affect many things down the line.

I will give another try in the next few days. I was doing the additional padded_mol = Chem.AddHs(padded_mol, addCoords=True) which may be causing the difficulties. Still interested in working on this

Changed the logic in _build_padded_mols to reproduce exactly the index mapping between padded molecule and rdkit molecule as develop.
Before 7ec6b66, the code was working for all but that part in export_flexres.py that uses index_map.
The latest version (7ec6b66) should be working.

Before the change, atoms added during the making of padded_mol were missing positions. After the change, the padded_mol has all hydrogens with conformations. This can be visually verified with the example file test/polymer_data/disulfide_bridge_in_adjacent_residues.pdb:

from meeko import Polymer, ResidueChemTemplates, MoleculePreparation
from rdkit import Chem

mk_prep = MoleculePreparation()
residue_chem_templates = ResidueChemTemplates.create_from_defaults()

input_file = "disulfide_bridge_in_adjacent_residues.pdb"
with open(input_file, "r") as f:
    pdb_string = f.read()

polymer = Polymer.from_pdb_string(
    pdb_string,
    residue_chem_templates, 
    mk_prep)

assert isinstance(polymer, Polymer), "Polymer object not created correctly"

for res_id in polymer.monomers:
    if res_id in ["G:187", "G:188"]:
        padded_mol = polymer.monomers[res_id].padded_mol
        assert padded_mol is not None, f"Padded molecule for residue {res_id} is None"
        
        writer = Chem.SDWriter(f"padded_mol_{res_id}.sdf")
        writer.write(padded_mol)
        writer.close()