2008/4/29 <Charles at schwieters.org>:
>
> Hello Jie-rong--
>
> I would first look at getting appropriate ensemble size. Recall, you
> have two sets of ensembles, one for the chromophore, and another
> describing your protein structure. You should probably first increase
> the size of the chromophore ensemble until the results converge - to get
> the appropriate size. Then do the same thing with your structure
> ensemble.
>
>
>
> Dear Charles,
The averaged Q-factors didn't change with the increasing conformers of
spin-label sites ("ALTs"), so I started the calculation with multi-conformer
proteins.
In the case of 5-conformer (5 identical polypeptides randomized at high
temperature without interacting each other) calcualtion, lots of violations
came out after "pins.showRestraint(0)". (The "finalTime" and "maxCycle" in
the cooling dynamics have been changed to 1.0ps and 500 respectively.)
If all the "RAMA" related terms are commented out, there is few "VIOLATED"
message after "pins.showRestraint(0)" in a 5-conformer calculation, while
the outcome structures are of course very bad (phi-psi angles are not in
protein prefered Ramachandra plot). It seems that the "RAMA" potential plays
a more important role than "PRE" potential during SA, is that corrent? I've
tried to change force constants ("ini_pre" "ini_rama" "fin_pre" "fin_rama")
in different ranges, but it didn't work. Is there any way to solve this
conflict? Thanks a lot!!!
Besides, here are some other questions:
The following message shown on x-term window and calculation stopped usually
at 58th or 59th structure (loop); the loop in script was sep up as 64.
####
xplor(22755) malloc: *** vm_allocate(size=754978816) failed (error code=3)
xplor(22755) malloc: *** error: can't allocate region
xplor(22755) malloc: *** set a breakpoint in szone_error to debug
Exception exceptions.AttributeError: "'NoneType' object has no attribute
'settrace'" in <bound method XplorSimulationPtr.__del__ of <C
XplorSimulation instance at _80312103_p_XplorSimulation>> ignored
####
What's the reason for this?
The "funType" (e.g. "harmonic") is not specified in the script. Is it
because that the type of penalty function is defaulted? (not described in
the prePot manual as well.)
How many "npc" should be set-in? My starting structure contains 8 cycteine
mutation cites, while the restraint tables (e.g. k6c.tbl, s20c.tbl...) are
read in one by one. What's the role of "npc" in calculation?
Best wishes,
Jie-rong
-------------- next part --------------
An HTML attachment was scrubbed...
URL:
http://dcb.cit.nih.gov/pipermail/xplor-nih/attachments/20080521/f496b203/attachment.html
