Alex Morehead

Hi, @greglandrum. Thanks for your quick reply. The following function is one I am using to construct `Chem.Mol` objects from (`.xyz`-like) 3D coordinate and element type input NumPy arrays. When...

Thanks, @bp-kelley! I'll give it a try

@bp-kelley, sadly, it doesn't look like that fixed the issue. I'm still seeing logs like the following being printed to my terminal when I call RDKit's bond determination functions: ```bash...

Help would be much appreciated. Otherwise, these logs will be printed each training step of an [open-source version of AlphaFold 3](https://github.com/lucidrains/alphafold3-pytorch). :upside_down_face:

@bp-kelley, is it possible to `pip install` the current state of the main branch to use this fix? If so, which command should I use? Just `pip install git+https://github.com/rdkit/rdkit.git@main` once...

OK. This might make it even trickier, since I have to install RDKit using `conda` for compilation-related performance reasons. How often is the `conda` package updated? Same as the UGM...

Thanks, @bp-kelley! I take it I'll be able to use this bug fix with the `2024_03_6` `conda` version of RDKit once it's released in early September?

@ptosco, sadly, it doesn't look like that worked. I'm still seeing the logs coming from `DetermineMonds()` even when redirecting the buffer manually.

@ptosco, an example of a SMILES string I'm giving to `rdDetermineBonds.DetermineBonds()` via ```python charge = Chem.GetFormalCharge(mol) with StringIO() as buf: with redirect_stderr(buf): # redirect RDKit's stderr to a buffer to...

It's definitely a PDB file. This `complex.pdb` file corresponds to the protein-ligand-nucleic acid complex structure which is denoted as target H1171v1 in CASP15 (and corresponds to PDB ID: 7PBL (https://www.rcsb.org/structure/7pbl)).