google-deepmind / google-deepmind/alphafold
Only use Structure module to predict structure with given representations
Open
- Dominant language
- Python
- Stars
- 14.9k
- Forks
- 2.9k
- PR merge metrics
- No merged PRs in 30d
Description
Hi,
I think it is possible to only use structure module to predict final structure when I have some intermediate representations, single and pair representations, so that I can bypass the Evoformer part. But is it easy to do so with a single call of structure module class? Like I load those representations from files, and I call the structure module class in your ```alphafold/model/folding.py```, with the input of those representations. If not, what else do I need to do?
Thanks!
Contributor guide
Assessment
This issue has not been assessed yet.