VMD-L Mailing List

From: Vermaas, Joshua (Joshua.Vermaas_at_nrel.gov)
Date: Thu Jun 06 2019 - 17:33:14 CDT

Hi Elizabeth,

Stride is can be fragile, and can give no output on some input structures. What I think is going on is that when you did this originally, you loaded the pdb, changed your representations (this is when Stride would have been called), and *then* loaded the trajectory file. In a visualization state, by default all the molecular loading file commands happen first, and then the representations are processed second, when stride is called on the active frame (usually the last one when loading from a visualization state). In this last frame configuration, Stride is apparently bailing out, resulting in no secondary structure information being read into VMD and the cartoon representation turns into spagetti. What I would try is to add these two lines to the end of your visualization state:

animate goto 0
mol ssrecalc top

The first line changes the active frame to the first one rather than the last, and then recalculates the secondary structure at that point, where Stride is known to work.

-Josh

On 2019-06-06 15:37:05-06:00 owner-vmd-l_at_ks.uiuc.edu wrote:

Hello, VMD users.

I am somewhat new to VMD (3 weeks). I have a trajectory of 5013 frames which I am viewing on an Ubuntu computer.
When I load the trajectory (xtc) into the molecule, which I load with a pdb, I can play the trajectory and everything works fine.
If I save the visualization state and open it again later, secondary structure is not recognized. In New Cartoon, the protein looks like gray spaghetti. All of my representations load fine, but even sequence viewer does not show any secondary structure.

This is the error I receive:

ERROR) Unable to find Stride output file: /tmp/fileYel4Lr
ERROR) Stride::read_stride_record: unable to read output file from Stride
ERROR) Call to Stride program failed.

I have made sure that Stride exists and the right path is in my bashrc.

I can open visualization states with my pdb, or pdbs and smaller trajectories (such as the files from the tutorial) and secondary structure is recognized properly.

I have told vmd to write stride SS files to ~/.vmdtmp/ rather than /tmp which should help if the issue is having sufficient memory.

Unfortunately, none of this helped. I have reduced my trajectory and pdb to only protein, GTP, and Mg and that works fine, but in some cases I may need to consider interactions with solvent and membrane. Would anyone be so kind as to offer any suggestions of what I can do?

Thank you,
Elizabeth