The difference between 0.sdf and 0_shifted.sdf

[Kartinaa]

Dear authors,

Thank you for sharing your wonderful work—it’s really intriguing. I’d like to understand the difference between the 0.sdf and 0_shifted.sdf files after I sampled molecules for a customized pocket. Were any post-processing methods applied to generate the 0_shifted.sdf file?

Thank you for your time, and I’d really appreciate any clarification you can provide.

Best, Bo

[Kartinaa]

Also, I encounted an error when trying to sample molecules from Crossdock dataset. I have downloaded and processed Crossdock dataset, the sampling command is python scripts/sample.py --outdir ./outputs --config ./configs/sample/sample_MolDiff.yml --batch_size 32 --pocket_dir CrossDocked_test --num_pharma_atoms 20 --clash_rate 0.1 and the error I had is File "MolSnapper/scripts/sample.py", line 121, in <module> with open(os.path.join(args.pocket_dir, 'filetered_uuid.json'), FileNotFoundError: [Errno 2] No such file or directory: 'CrossDocked_test/filetered_uuid.json' Many thanks!

[YaelZiv]

Hi Bo, the only difference between the 0.sdf and 0_shifted.sdf files is in their centering. The 0.sdf file is centered at zero (0,0,0), while the 0_shifted.sdf file has been shifted back to the original center of the pocket. No other post-processing methods were applied—this shift simply repositions the molecules in relation to the original constrains' spatial coordinates.

In addition, the file 'filetered_uuid.json' is in data/crossdocked/ please move it to your data directory best, Yael

[Kartinaa]

Thanks so much Yael, it helps a lot!