-
Notifications
You must be signed in to change notification settings - Fork 273
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
residue.is_nucleic and 'nucleic' in atom selection DSL #1398
base: master
Are you sure you want to change the base?
Conversation
If I had to guess, it was probably a combination of
|
Thanks -- I thought it might just be that the MDTraj user/dev community is more focused on proteins than nucleic acids, so this hasn't been an itch anyone else wanted to scratch. This PR is probably worth continuing, then. Any thoughts on the question about using unofficial (and possibly conflicting) residue names? As for the list, I think ParmEd has a pretty solid list of nucleic acid residues and common aliases/variants for them (though missing inosine): |
@dwhswenson: is this still relevant? I'm happy to hit "merge". |
It's nearly ready. I've been meaning to ask what you think should be included as "nucleic", but got distracted by other things. In short, there are several possibilities to consider (beyond the obvious and expected residues):
I had some discussion on this with the ParmEd folks in ParmEd/ParmEd#1047. I put some detailed discussion in a gist. (Since I'm having trouble getting GitHub to load that gist right now, here's the notebook in nbviewer.) I've designed what I've included to make it relatively easy (though not trivial) for users to customize the behavior here. The notebook in that gist shows how to generate a |
Honestly I don't really know anything about the conventions for nucleic acid names, so I'm not really in a position to review much. |
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
9087f56
to
29dd0ec
Compare
9ca5ba0
to
3adc2d8
Compare
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
Is there anybody we could recruit to give a review of this, even if it's only a brief skim? I'm in less of a position to speak with authority on nucleic acid naming conventions. I'm open to just merging this as-is ... has anything change since your comment in May 2019? |
Mainly I keep dragging this "forever PR" along, updating it to master every time it gets marked stale, but not bringing it to conclusion. Sorry about that. I think my primary thoughts on the matter are exposed a little in the gist mentioned above. There are several options there; I think my preference is now leaning toward "let's default to something reasonably minimal, but provide users with an easy way to customize." In the 4 years (programming gods forgive me) since I opened this PR, I'm hoping that the non-standard Amber-specific names that I used to see are used less frequently. My vote would be for boring standard residues as default "nucleic" (so keep the It's been a while since I've checked that my janky awk-based parsing in that notebook works, although it should. I'd be very open to an improvement on that ugly mess. (I went through an awk phase in the mid 2010s that apparently hadn't been fully cured before that gist. Sorry.) |
This issue has been automatically marked as stale because it has not had recent activity. It will be closed if no further activity occurs. Thank you for your contributions. |
This is a really simple PR aiming to add support for
residue.is_nucleic
and'nucleic'/'is_nucleic
as special query strings (like'protein'
) in the atom selection language.Main question: Is there any reason not to do this? It seems like such low-hanging fruit, and it was half-implemented, so I wondered if some significant concern I'm not aware of was preventing it.
It looks like the last time anything was done with this was way back in #594, when nucleic was officially treated as "not supported" by commenting out several lines that were still comments in the code.
A few specific questions to ask: