Hole analysis on trajectory converted to DCD

[Daniel Burns]

I have a bunch of xtc trajectories that I am using the hole2 analysis on after converting the xtc files to dcd with mdanalysis.

It seems to work fine but the guide says:

'HOLE can not read DCD trajectories written by MDAnalysis, which are NAMD-style. ignored.'

Should I be concerned about my results?

Thank you,

Dan

[Oliver Beckstein]

Hi Dan,

It depends on how you’re doing your analysis. If you use HoleAnalysis on a universe that contains the trajectory then the comment regarding HOLE and DCD does not matter: HoleAnalysis extracts frames from any trajectory to temporary PDB files and feeds them to HOLE, frame by frame. (In this case you can just use your XTC.)

If you use hole(…, dcd=DCD) then you are asking HOLE itself to read the trajectory. In this case I’d be careful and check that the output is what you expect. It *might* work, but I don’t remember that I ever got that to work.

Oliver

--

Oliver Beckstein (he/his/him)

email: orbe...@mdanalysis.org
twitter: @orbeckst

GitHub: @orbeckst

MDAnalysis – a NumFOCUS fiscally sponsored project

https://www.mdanalysis.org/

[Daniel Burns]

I see, thank you.

I've run it and got a number of bad profiles (but mostly great ones, which is awesome!) in results.['profiles'].['radius'].

 I'm going to rerun it with the cpoint specified.  

I've aligned the trajectory based on the atom selection that defines the pore so the center point is almost exactly the same between each frame.

Do I just leave that one specified point to refer to all frames across the trajectory or is there a way to have cpoint updated each frame?

Thanks again!

Dan