Esmfold #114
Esmfold #114
Conversation
|
Check out this pull request on See visual diffs & provide feedback on Jupyter Notebooks. Powered by ReviewNB |
docs/CHANGELOG.md
Outdated
| @@ -14,7 +27,8 @@ The CHANGELOG for the current development version is available at | |||
|
|
|||
| ##### New Features | |||
|
|
|||
| - Added support for [AlphaFolds 200M+ structures](https://www.deepmind.com/blog/alphafold-reveals-the-structure-of-the-protein-universe) via `PandasMmcif().fetch_mmcif(uniprot_id='Q5VSL9', source='alphafold2-v3')` and `PandasPdb().fetch_pdb(uniprot_id='Q5VSL9', source='alphafold2-v3')`. | |||
| - Added support for [AlphaFolds 200M+ structures](https://www.deepmind.com/blog/alphafold-reveals-the-structure-of-the-protein-universe) via `PandasMmcif().fetch_mmcif(uniprot_id='Q5VSL9', source='alphafold2-v3')` and `PandasPdb().fetch_pdb(uniprot_id='Q5VSL9', source='alphafold2-v3')`. (Via [Arian Jamasb](https://github.com/a-r-j), PR #[102](https://github.com/rasbt/biopandas/pull/102/files)) | |||
There was a problem hiding this comment.
Awesome PR! Btw in the changelog here, it should mention the new ESMFold support alongside alphafold, correct?
There was a problem hiding this comment.
I suppose we could, but I think the AF2 support has already been released so kept them separate
There was a problem hiding this comment.
Ah sorry, I didn't mean to change the comment but to add the ESMfold note below.
| """Test retrieving a structure from ESMFold.""" | ||
| sequence = "MTYGLY" | ||
| res_ids: Set[str] = {"A:MET:1", "A:THR:2", "A:TYR:3", "A:GLY:4", "A:LEU:5", "A:TYR:6"} | ||
| ppdb = PandasPdb().fetch_pdb(sequence=sequence) |
There was a problem hiding this comment.
Should this include ESMfold as the source? Just as a contingency for when new versions of ESMfold come out. Or if alternatives come up. (But I suppose right now fetching an AlphaFold structure via sequence is not possible via AlphaFold, right?)
There was a problem hiding this comment.
Yeah, no API for AF2 inference AFAIK
There is a version arg for the ESMFold function but it remains to see exactly how the API will change for subsequent releases so may be okay to leave for the time being.
|
@a-r-j It looks all good to me! Just added I noticed that the PDB jupyter notebook is corrupted though. Couldn't open it on my computer, and I also see that that it's the same for Gh rendering: https://github.com/rasbt/biopandas/blob/4515e39861a422c10a8aff7677ecd5aacb031508/docs/tutorials/Working_with_PDB_Structures_in_DataFrames.ipynb Do you have a working version on your computer still? Othewise I will restore it from an earlier commit. |
|
Oh gosh, let me see what I can do. Keeping notebooks under version control is always a nightmare. |
|
Seems I hadn't resolved all the conflicts when merging. Should be fixed now :) |
|
Thanks, I will try to check that out today! I will also take care of the docs. But just if you are curious, there are a few notes here: https://biopandas.github.io/biopandas/CONTRIBUTING/#notes-for-the-developers PS: I transferred the repo to a new BioPandas org. Looks like everything worked pretty smoothly so far |
|
I think you undid my commits where I added |
|
Oh gosh, my bad. Looks like I showed up to amateur hour 😓 I can't find your changes in the commit history - any chance you have a local copy? |
|
No worries, thinks happen. Unfortunately, I destroyed them when I pulled your recent branch ... arg |
Description
Adds support for retrieving predicted structures from ESMFold.
Pull Request Checklist
./docs/sources/CHANGELOG.mdfile (if applicable)./biopandas/*/testsdirectories (if applicable)biopandas/docs/sources/(if applicable)PYTHONPATH='.' pytest ./biopandas -svand make sure that all unit tests pass (for small modifications, it might be sufficient to only run the specific test file, e.g.,PYTHONPATH='.' pytest ./biopandas/classifier/tests/test_stacking_cv_classifier.py -sv)flake8 ./biopandas