You signed in with another tab or window. Reload to refresh your session.You signed out in another tab or window. Reload to refresh your session.You switched accounts on another tab or window. Reload to refresh your session.Dismiss alert
We'd like to support Protein-Ligand Interaction Graphs eg as shown in this image from this paper.
Ideally we can keep this composable under the hood. E.g. we could take a protein graph computed via the graphein.protein API and molecule graph computed with the graphein.molecule API and merge them into one graph (and separate them into two).
The hardest aspect of this is remove non-ligand and non-protein HETATMs from the PDB records. There will be metal ions, crystallographic adjuvants and co-factors that we may wish to exclude. We can maintain common sets of these in our lists of constants and provide an access point for users to supply their own.
A starting point can be this gist from Pat Walters.
The text was updated successfully, but these errors were encountered:
Follow on issue for #155
We'd like to support Protein-Ligand Interaction Graphs eg as shown in this image from this paper.
Ideally we can keep this composable under the hood. E.g. we could take a protein graph computed via the
graphein.protein
API and molecule graph computed with thegraphein.molecule
API and merge them into one graph (and separate them into two).The hardest aspect of this is remove non-ligand and non-protein HETATMs from the PDB records. There will be metal ions, crystallographic adjuvants and co-factors that we may wish to exclude. We can maintain common sets of these in our lists of constants and provide an access point for users to supply their own.
A starting point can be this gist from Pat Walters.
The text was updated successfully, but these errors were encountered: