Menu
▾
▴
#252 Add support for FAHClient viewerTop.json format
From what I'm seeing FAHClient hasn't used the current.xyz format in a while. It now produces a viewerTop.json and several viewerFrame00.json files. Being able to load these file types instead would make for a seemless live browser based view of the FAHClient.
Related
Please supply some sample files.
On Mon, May 4, 2020 at 2:37 PM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
I believe these are the only files needed to reconstruct the protein visual. The viewerFrameXX.json files seem to show a progression of growth of the protein throughout the simulation. You should only need the viewerTop.json file and any one viewerFrameXX.json file to recreate a frame. I've included both the initial frame and my most recent frame.
Q: Do we need to/want to be able to compare two or more frames? or are
these always trajectories -- that is, we just want to quickly load new atom
positions.
I like this format. Should be a simple fix.
On Tue, May 5, 2020 at 11:17 AM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
I don't know if we necessarily need to be able to compare them side-by-side, but I do like the idea of being able to quickly load them in succession to create an animation of the protein growth. That may be able to be accomplished with just scripting on the client though.
Glad to hear you like the format and that it should be a simple fix!
Last edit: Joey Paris 2020-05-05
How does one generally get these files? Results of a calculation? Can I
presume that both files are available locally? Can Jmol request them both
individually? As a set? Would we first have the "top" and then if frames
are changed only need to load Frame[n]?
On Tue, May 5, 2020 at 11:17 AM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
These are generated in the
workdirectory by the FoldingAtHome client (FAHClient). On Mac these files are located in/Library/Application Support/FAHClient/work/XX/where XX is a 0 padded 2-digit number, that seems to depend on what projects are being actively worked on by the FAHClient although I'm not 100% positive that's how that number is generated. The path on Linux machines is/var/lib/fahclient/work/XX/. I am unsure about the Windows work directory path.Both files should be available locally, as long as the FAHClient is actively working on a simulation.
From my understanding (albeit limited), You only need the
viewerTop.jsonfile once, and then you can get the different frames just by requesting the specificframe[n]file.Last edit: Joey Paris 2020-05-05
OK, I propose:
1) You can load the top file alone. It will just show one dot (all the
atoms at 0 0 0).
2) A second load using LOAD TRAJECTORY to load the coordinates will create
the actual model.
3) Subsequent LOAD TRAJECTORY will modify the coordinates only.
4) Alternatively, you can use LOAD with the "file1+file2" option to load
both at once and display a model, as with PDB files.
5) Using LOAD APPEND file1+file2 you can see to models at the same time (in
different frames unless set appendNew is FALSE)
I should be able to have this for you tomorrow.
That sounds like a perfect solution to me! I honestly was not expecting to
get an answer this quickly let alone a solution. I really appreciate it!
Thank you and let me know if there’s any more information I can give you!
Last edit: Joey Paris 2020-05-06
How's this?
new feature: FAHReader FAH:: reads FoldingAtHome Top.json and
FrameNNN.json files
-- FAH:: is required for Frame files but not the Top file.
-- automatic calculation of structure using DSSP
-- Both files are required; two load options:
to two files
coordinates
associated ViewerTop.json assumed present
-- Subsequent calls to LOAD XYZ will replace coordinates and recalculate
DSSP only.
On Tue, May 5, 2020 at 6:25 PM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
Q: What are the "units" line options?
On Wed, May 6, 2020 at 9:40 AM Robert Hanson hansonr@stolaf.edu wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
Based on the FAHViewer client (
https://github.com/FoldingCommunity/fah-viewer/blob/a8d0f20d6cc1944f20d3cdd71885c1fc177773c8/src/fah/viewer/Trajectory.cpp)
if
I am reading it correctly, the options
NMorNANOMETERSgivethe model a scale of 1, and the options
A,ANGSTROM, orANGSTROMSgive themodel a scale of 10. I don't know enough about protein visual
structures to know where that scale is applied, but hopefully that's enough
information for you to move forward!
Also, it appears if no units are supplied it defaults to a scale of 10
Last edit: Joey Paris 2020-05-06
OK, I will set that up. Right now it will only accept ANGSTROMS
Please test and get back to me the Jmol.jar file in
https://sourceforge.net/p/jmol/code/HEAD/tree/trunk/Jmol/test/jmol-jar/
On Wed, May 6, 2020 at 10:05 AM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
please send a NANOMETER example if you can.
On Wed, May 6, 2020 at 10:07 AM Robert Hanson hansonr@stolaf.edu wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
one more update
https://sourceforge.net/p/jmol/code/HEAD/tree/trunk/Jmol/test/jmol-jar/
This one does a quick bond distance check to ensure that the units match
the actual structure (which they did not, in this case!)
On Wed, May 6, 2020 at 10:35 AM Bob Hanson hansonr@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
That's looking much more appropriately scaled! The console help text is a nice addition as well.
Last edit: Joey Paris 2020-05-06
Yes. So that ViewerTop file is wrong. These Frame files are in NM.
On Wed, May 6, 2020 at 11:28 AM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
Interesting. I wonder if that's an underlying problem with the FAHClient since I got those files straight from there. Here's some new files from what my computer is currently working on so we can see if it's a consistant inaccuracy.
I also probably should have asked earlier, but once this update goes live the JSmol package will automatically be updated as well, correct?
No question that these are in NM.
[
[0.844847, -2.192282, 0.472762],
[0.824088, -2.148159, 0.561211],
[0.782002, -2.264074, 0.439637],
[0.926617, -2.244784, 0.500298],
[0.872098, -2.087131, 0.373569],
[0.948068, -2.021726, 0.416368],
[0.924359, -2.139544, 0.240121],
[0.845438, -2.193618, 0.187885],
[1.001068, -2.216059, 0.252044],
WAY too small numbers for Angstroms.
Good news is that I have code now that allows Jmol to auto-scale these
regardless of what "units" reads. The user gets a notification:
FAHReader found NANOMETERS based on bonds[0].length=0.15197037160396576
Means they have all been scaled up by a factor of before leaving FAHReader.
update - yes, I think we're ready for that now.
On Wed, May 6, 2020 at 12:15 PM Robert Hanson hansonr@stolaf.edu wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Awesome! Thanks again for the super quick turn around on this. I will open a ticket on the FoldingAtHome github about the output unit being incorrect.
I have changed the message to:
FAHReader CORRECTION: Top.json file units=ANGSTROM
but we found NANOMETERS based on
bonds[0].length=0.15197037160396576
So you can send that to them, if you want. Here is the updated package:
*Looking for the latest version? * Download Jmol-14.30.2-binary.zip (72.6
MB) https://sourceforge.net/projects/jmol/files/latest/download
BTW, code for this is at
https://sourceforge.net/p/jmol/code/HEAD/tree/trunk/Jmol/src/org/jmol/adapter/readers/simple/FAHReader.java
On Wed, May 6, 2020 at 12:25 PM Joey Paris joeyparis@users.sourceforge.net
wrote:
--
Robert M. Hanson
Professor of Chemistry
St. Olaf College
Northfield, MN
http://www.stolaf.edu/people/hansonr
If nature does not answer first what we want,
it is better to take what answer we get.
-- Josiah Willard Gibbs, Lecture XXX, Monday, February 5, 1900
Related
Feature Requests: #252
Looks like everything is working as expected! Played around with the JSmol version a bit and it's working great. Thanks again for all your help with this.
I also got that ticket opened on GitHub so hopefully someone there can address it!
Now on to making the background transparent during the live render, but I'm trying to figure that out myself.
I ran through the three sample ways to load in coordinated like you mentioned and all generated basically this visual! There wasn't much change between the frames, but I'm assuming that's just because the protein didn't actually change that much.
I also ran it with what my computer is currently folding and they seemed to have similar shape. In both FAHViewer and Jmol. The only thing I couldn't figure out was how to decrease the size of the atoms, but I'm pretty sure that was just user error.