-
Notifications
You must be signed in to change notification settings - Fork 27
Conversation
Ignore Mac .DS_Store file
Added bug report tempalte
Update issue templates
Added feature request issue template
set export to False
Development
Bumps [freesasa](https://github.com/mittinatten/freesasa) from 2.0.3.post7 to 2.0.5. - [Release notes](https://github.com/mittinatten/freesasa/releases) - [Changelog](https://github.com/mittinatten/freesasa/blob/master/CHANGELOG.md) - [Commits](https://github.com/mittinatten/freesasa/commits) Signed-off-by: dependabot-preview[bot] <support@dependabot.com>
Bump freesasa from 2.0.3.post7 to 2.0.5
Updates the requirements on [torch](https://github.com/pytorch/pytorch) to permit the latest version. - [Release notes](https://github.com/pytorch/pytorch/releases) - [Commits](pytorch/pytorch@v0.1.1...v1.4.0) Signed-off-by: dependabot-preview[bot] <support@dependabot.com>
Pull Request Test Coverage Report for Build 784
💛 - Coveralls |
Shall we use residue center for all residues instead of for residues with missing atoms? |
it was at the center first but then we decided to move to CB ... |
Test realign
…to skip_empty_features fix test of the realign
I suggest we leave the features like that for now, This PR should fix the issue but we have then different default for different features when the PDB is faulty ... |
373ccda
to
f2021cb
Compare
deeprank/features/FullPSSM.py
Outdated
@@ -212,6 +212,14 @@ def get_feature_value(self, cutoff=5.5): | |||
f"{self.mol_name}: The following interface residues have " | |||
f" no pssm value:\n {ctc_res_wo_pssm}" | |||
) | |||
elif len(pssm_res_set.difference(ctc_res_set)) > 0: |
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The pssm_res_set
contains pssm for all residues, and ctc_res_set
consider only interface residues, so this condition should always be True.
Also pssm_res_wo_ctc = pssm_res_set.difference(ctc_res_set)
misstreats the non-interface residues as residues with missing atoms.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
The issue is that some of the res are missing in xyz_dict therefore leading to a keyError on line 229. I think we should create a get_residue_centers
either in DeepRank or pdb2sql. There we could specify to get either CB, or CA if CB is missing or take the mean of the residue xyz if CA and CB are missing. The only issue is the we would have to loop over the residue and it will be a bit slow for large complexes ....
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I think we need a data preprocess step in DeepRank to avoid these problems. We can could add a PDB validator in pdb2sql
.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
that's a good idea ... but I wouldn't know how to do it
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
I've introduced a get_residue_center in FetureClass I think we can live with that for now ....
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks, the get_residue_center
looks good.
There was a problem hiding this comment.
Choose a reason for hiding this comment
The reason will be displayed to describe this comment to others. Learn more.
Thanks, the changes look good.
handles missing CA/CB atoms in residue based features