search

microsoft / foldingdiff

Diffusion models of protein structure; trigonometry and attention are all you need!
https://www.nature.com/articles/s41467-024-45051-2
MIT License
518 stars 57 forks source link

ERROR 404: Not Found when I download the AlphaFold files #1

Closed sirius777coder closed 2 years ago

sirius777coder commented 2 years ago

Great Work. I meet some small questions, when I use your script to download the AlphaFold data, it tells me that ERROR 404: Not Found. May be there is something wrong with the AlphaFold database path.

wukevin commented 2 years ago

We haven't fully set up the AlphaFold datasets! We are currently focused on the CATH dataset and may eventually extend to AlphaFold dataset (or others) in the future.

sirius777coder commented 2 years ago

Thanks for your quick reply. Does models with different pdb data have different performance?

sirius777coder commented 2 years ago

And when I download the CATH data , it said : md5sum: cath/cath-dataset-nonredundant-S40.pdb.tgz: no properly formatted MD5 checksum lines found. It is correct?

wukevin commented 2 years ago

We haven't done an extensive evaluation on different datasets to be able to say which dataset produces optimal results. I'd intuitively expect that the more data you give the model to learn from, the better the end results.

And yes the error message is expected.