datamol
datamol copied to clipboard
datamol-io
Molecular Processing Made Easy.
Proposed change to conformer force field from UFF to MMFF94s_noEstat, inspired by Openeye's Omega force field. In addition to FF change, added filters (ewindow, eratio), to allow high-energy conformers to...
Hi, Thanks for the nice package. Looking at _FP_DEFAULT_ARGS in to_fp, it has useFeatures=True for 'ecfp' class. If I'm not mistaken, this leads to calculation of FCFP rather than ECFP...
According to the datamol documentation and code, the enforce_subs parameter of dm.scaffold.fuzzy_scaffolding should be a list of str (presumably SMILES or SMARTS). However, passing in a list of str gives...
See https://github.com/datamol-org/datamol/blob/013d93012abb7fb309a382d8b3eedaca4c2f4425/datamol/align.py#L107 We should not modify the global behaviour. The best is probably to put back the original value if it has been modified. This is not ideal in when...
- it does not seem to respect the 2D position of the input template. - Under the hood it uses `rdMolAlign.AlignMol` which seems to be for 3D. That could explain...
``` scf.UpdatePropertyCache() Chem.GetSymmSSSR(scf) ``` this is not done by default and can sometime lead to `Pre-condition Violation - RingInfo not initialized` errors.
The current datamol unique id looks like `6cb8a819d03a38a6ed4771d1bbfd899c`. In the future, we might want to rely on other type of hashing such as the soon-to-come in rdkit: https://github.com/rdkit/rdkit/pull/5360 It might...
`dm.scaffold.fuzzy_scaffolding` is quite a powerful function but its output is often hard to understand and also process for downstream task. We could keep backward compat by keeping `dm.scaffold.fuzzy_scaffolding` and propose...
This would simplify things when working in the context of querying a rest api where the json response contains a list of molecules. Quick REST API examples: - Molport -...
