AMBER Archive (2006)

Subject: Re: AMBER: igb=2 and "bad atom type" conflict

From: Evan Kelly (ebkelly_at_ualberta.ca)
Date: Thu Jul 06 2006 - 17:01:39 CDT

As per the discussion that Scott and I have been having (see below),
can anyone elaborate on the meanings of the parameters x(l170-1+i), x
(l175-1+i), etc. in mdread.f? In particular, I am looking to modify
the code to handle Fluorine.

Thanks!

On 6-Jul-06, at 3:44 PM, Scott Pendley wrote:

> On 6-Jul-06, at 9:47 AM, Scott Pendley wrote:
>
>> You are on track. I would leave lines 3-7 (I170-I185)set to
>> 0.00000 unless you actually know the parameters. If I remember
>> correctly remember they related to a complex equation for
>> calculating surface area with overlapping atoms. You will
>> probably want to alter sections for both igb=1 and igb=2.
>>
>> On 7/5/06, Evan Kelly < ekelly9_at_gmail.com> wrote:
>> Hi Scott,
>>
>> Thanks for the tip. Just to make sure I understand the code and
>> the data storage in AMBER (I've never had to modify this code
>> before):
>>
>> 1) For the parameters I have used, I need to add F parameters in
>> the section starting at line 1054 of mdread.f, "if ( gbsa == 1 )
>> then"?
>>
>> 2) I should add a line like:
>>
>> else if (atype == 'F' .or. atype == 'f') then
>> x(l165-1+i) = 1.47d0 + 1.4d0
>> x(l170-1+i) = 0.68563d0
>> x(l175-1+i) = -0.1868d0
>> x(l180-1+i) = -0.00135573d0
>> x(l185-1+i) = 0.00023743d0
>>
>> into the series of elseif statements? The l165 parameter is the
>> vdw radius "augmented by 1.4 Ang.". However, I have no idea what
>> the other parameters stand for or what they should be for F. Any
>> suggestions?
>>
>> 3) I should only need to recompile the sander code, by "make
>> serial" in $AMBERHOME/src/sander/, right?
>>
>>
>> Please let me know if I'm on the right track. Thanks!
>>
>>
>>
>> On 5-Jul-06, at 2:22 PM, Scott Pendley wrote:
>>
>>> Hello Evan,
>>>
>>> You will need to modify the AMBER source code to enter fluorine
>>> parameters. I would go to $AMBERHOME/src/sander/mdread.f and add
>>> in parameters. Keep in mind the vdw radius for fluorine is 1.47
>>> Angstroms. Have fun and let me know if you need any help with this.
>>>
>>> Scott
>>>
>>> On 7/5/06, Evan Kelly < ebkelly_at_ualberta.ca> wrote:
>>> Hi Everyone,
>>>
>>> I am trying to run a sander calculation on a protein ligand
>>> complex. I used antechamber to create the prmtop and inpcrd
>>> files for the ligand. I am using the igb=2 option. I get the
>>> following error:
>>>
>>>> bad atom type: f
>>>
>>> here is my in file:
>>>
>>>> &cntrl
>>>> imin = 1,
>>>> drms = 0.0001
>>>> ntmin = 1,
>>>> ncyc = 50,
>>>> maxcyc = 100,
>>>> ntb = 0,
>>>> ntf = 1,
>>>> ntc = 1,
>>>> cut = 30,
>>>> ntr = 0,
>>>> igb = 2, saltcon= 0.1, rgbmax = 30, gbsa = 1,
>>>> ntpr = 1, ntwx = 100, ntwr = 100,
>>>> /
>>>
>>> Information of note:
>>>
>>> 1) a similar system that does not contain F atoms works perfectly
>>> fine.
>>> 2) I tried changing the 'f' atom types in the prmtop file to 'F'
>>> and the same problem occurred (bad atom type: F)
>>> 3) the error does NOT occur when I turn GB solvation off (igb = 0)
>>> 4) this was run on two different systems to the exact same error.
>>>
>>> Are there any suggestions or comments? Thanks in advance!
>>>
>>>
>>> ----------------------------------
>>> Evan Kelly
>>> ebkelly_at_ualberta.ca
>>>
>>>
>>>
>>
>>
>> --------------------------------
>> Evan Kelly
>> ekelly9_at_gmail.com
>>
>>
>>
>>
>
>
> ----------------------------------
> Evan Kelly
> ebkelly_at_ualberta.ca
>
>
>

----------------------------------
Evan Kelly
ebkelly_at_ualberta.ca

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu