[Chimera-users] Fitmap in batch mode

[ingvar]

Hello,

I am trying to run fitmap on a set of volumes on a server without a 
display.  I would like to collect the results of the top hits to store 
in a database – correlation coefficients, inclusion scores, and the 
transforms.  The only way I found to collect the fitting results is to 
save the session after displaying the listFits dialog.   The listFits 
dialog cannot be displayed when running Chimera in –nogui mode, 
otherwise this would have been a solution.

I tried to use a virtual display, with Xvfb.  I have used Xvfb with 
some java applications in the past.  When I use Xvfb with Chimera, it 
appears that Chimera do not want to start up properly, the process just 
hangs, i.e. no cpu usage.

Is there a way around this, or am I asking for an enhancement, e.g. a 
“writeListFile toFile” option.

I am on RHEL 6.2/6.3 if that matter.

Noticed that there is a local correlation option in the pipeline for 
Chimera 1.8, looking forward to try it out.

Kind Regards,
Ingvar

Typical python script:
from chimera import runCommand as rc
rc("open emd_5500.map")
rc("volume #0 level 1.34 transparency 0.5")
rc("open emd_2017.map")
rc("volume #1 level 0.58 transparency 0.5")
rc("fitmap #1 #0 search 200 metric overlap envelope true inside 0.2 
listFits false")
#rc("fitmap #1 #0 search 200 metric overlap envelope true inside 0.2")
rc("save session5500_2017")
rc("stop now")

Setup to run with Xvfb (in bash):
i=0
while [ -f /tmp/.X$i-lock ]
do
   i=$[$i+1]
done
Xvfb :$i &
pid=$!
export DISPLAY=:$i
chimera fit5500_2017.py
kill $pid