Thank you all for your replies and suggestions!
I ended up adding a small non zero vdw radius. The minimization went alright, so did heating and initial equilibration, no collapse or significant structure distortion observed.
I experience some problems with npt equilibration, but right now I have no reason to connect them with (ho) atom type behavior (...-PO3H2).
This (ho) atom type problem has been discussed in the community before,
so I attached a few links in case somebody finds them useful.
http://code.google.com/p/acpype/wiki/FAQ
http://structbio.vanderbilt.edu/archives/amber-archive/2003/1967.php
http://archive.ambermd.org/200607/0109.html
Matthew Terrier suggested to modify (ho) vdw params in here:
http://archive.ambermd.org/201110/0119.html
I could not find the actual publication behind this modification, it could be a conference poster or an oral presentation.
It seems many people modify (ho) vdw params to the ones for other types of hydrogen.
There is also additional (ho) vdw problem discussion in the lmod simulation protocol manual.
With best regards, Maria.
On Jul 12, 2012, at 8:15 PM, Bill Ross wrote:
> I would create another type of H (e.g. HX), like HO in all respects but vdw.
>
> Bill
>
> Maria Minakova <mariabiophysics.gmail.com> wrote:
>
>> Oh I see. I will think about other options, though concern is that this particular atom type is designed for OH hydrogen, while others might have different bond, angles and etc. parameter values.
>> Would you suggest to modify the definition of ho to give it nonzero wdv radious via frcmod?
>>
>> With best regards, Maria.
>>
>> On Jul 12, 2012, at 7:20 PM, Bill Ross wrote:
>>
>>> If the HO type has 0 vdw radius, you may want to use a type that
>>> has > 0 radius to prevent it moving through the OH vdw.
>>>
>>> Bill
>>>
>>>>> I am minimizing a complex synthetic compound, which also has phosphonic acid groups attached to it:
>>>>> OH
>>>>> |
>>>>> R - P = O
>>>>> |
>>>>> OH
>>>>> The automatic atom type assignment (gaff) put (py) atom type for central P, (o) for double bond O, (oh) for O in OH fragment, and (ho) for H in OH fragment.
>>>>> I did not put any restrictions for minimization, and within fist 1000 steps hydrogens collapsed onto neighboring oxygens, which is not supposed to happen.
>>>>> I wonder if there is something wrong with my parametrization or minimization script.
>>>>> Any ideas or suggestions?..
>>>>>
>>>>> additional info:
>>>>> Partial charges were assigned by mulliken method (resp didn't work well with the rest of this compound).
>>>>
>>>> _______________________________________________
>>>> AMBER mailing list
>>>> AMBER.ambermd.org
>>>> http://lists.ambermd.org/mailman/listinfo/amber
>>>
>>> _______________________________________________
>>> AMBER mailing list
>>> AMBER.ambermd.org
>>> http://lists.ambermd.org/mailman/listinfo/amber
>>
>> _______________________________________________
>> AMBER mailing list
>> AMBER.ambermd.org
>> http://lists.ambermd.org/mailman/listinfo/amber
>
> _______________________________________________
> AMBER mailing list
> AMBER.ambermd.org
> http://lists.ambermd.org/mailman/listinfo/amber
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Mon Jul 30 2012 - 14:30:05 PDT
