Re: [AMBER] How to make the picture of secstruct and the analysis of Hbond

From: Daniel Roe <daniel.r.roe.gmail.com>
Date: Mon, 18 Jul 2011 09:25:54 -0400

Hi,

> Estimated memory usage for this hbond call: 25.35 GB

Do you have this much memory available on the computer you are running
the ptraj hbond analysis on? If not, this is the reason for your
segfault. The estimate is based on the number of possible hbonds, as
well as the total # of frames that will be read in, so break up your
analysis into manageable chunks. E.g. if you have 8 GB of memory
available then reading in 500 sets at a time should be OK.

-Dan

On Fri, Jul 15, 2011 at 11:21 PM, qiao xue <xueqiaoup.gmail.com> wrote:
> Thanks very much, Roe.
>
> I am using the Amber 10 and Ambertool1.2. The bugfix is 10.0 bugfixes.
> And the compilers is Inter 11.1. The OS is linux.
>
> My system is made up of 127 residues. I did 5ns. There are 2500 frames
> in the md trajectory.
>
> the input and out file are attached.
>
> 2011/7/14, Daniel Roe <daniel.r.roe.gmail.com>:
>> Hi,
>>
>> On Wed, Jul 13, 2011 at 7:54 PM, qiao xue <xueqiaoup.gmail.com> wrote:
>>> When I am doing the Hbond analysis it often suffers the error
>>> "segmentation fault".
>>
>> Can you please post more details about your system so we can try and
>> reproduce the error?
>> - What version of AmberTools are you using, what bugfixes have you
>> applied, what compilers (and version) are you using and on what OS?
>> - How large is your system (# atoms), how many frames is your
>> trajectory, are box coords present? Ideally attach the topology file
>> and 1 coordinate frame or restart file.
>> - Post the exact ptraj input you are using, and optionally attach the
>> output you are getting.
>> Thanks!
>>
>> -Dan
>>
>> _______________________________________________
>> 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 18 2011 - 06:30:04 PDT
Custom Search