tauto_k in get_stable_tautomers
nadiahfl opened this issue · comments
Describe the bug
The tauto_k is said to output the top-k tautomers for each SMILES. When testing on the tutorial example sildnafil.smi (and other molecules), changing tauto_k in get_stable_tautomers function doesn't seem to change the number of output tautomers.
To Reproduce
Steps to reproduce the behavior:
- Go to the tautomer tutorial (example/tautomer.ipynb)
- set tauto_k=3 to see it generates 3 tautomers
- set tauto_k=2 to see it still generates 3 tautomers
System information:
- Operating System:
- Auto3D version: 2.2.9
- Python version:3.9.7
- RDKit version:2023.09.4
- PyTorch version:2.1.0
Additional context
Add any other context about the problem here.
Hello, thanks for the detailed description. I tried to reproduce the issue, but did get 2 tautomers when setting tauto_k = 2
.
A place susceptible to error is argument k
. Could you make sure that your k=1
in the options
function? Besides, I also attached my jupyter notebook below for your reference. Let me know if you have further questions.
tautomer.ipynb.txt
Thanks for your helpful reply. I had tauto_k=2 and k=1, and the smi_taut.smi file has three tautomers which I thought is supposed to be two for tauto_k=2. I probably misunderstood this as it looks that the stable tautomers are written at the very end.
I see where the problem is. The stable tautomers are stored in an SDF file, of which the path will be printed out in the terminal/console after Auto3D finishes running. There should be something like the following that specify the output SDF path in the terminal/console.
Begin to select stable tautomers based on their conformer energies...
Done.
The stable tautomers are stored in: /home/jack/Auto3D_pkg/example/files/sildnafil_20240329-144811-182857/sildnafil_out_top_tautomers.sdf