Re: Using ABF to explore the conformational space of a spin label attached to a membrane protein

From: Jrme Hnin (jhenin_at_ifr88.cnrs-mrs.fr)
Date: Thu Jan 20 2011 - 08:07:36 CST

On 19 January 2011 17:01, Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com> wrote:
> A quick question on subdividing a PES into smaller parts (when doing ABF):
> Should there be some overlap between the smaller regions in order to avoid
> the effects of the artificial wall constraint? Can such
> overlapping regions be put together using inputprefix?

If there is overlapping data, it will be handled properly, but the
overlap is not necessary.
Jerome

> On Wed, Jan 19, 2011 at 16:23, Jérôme Hénin <jhenin_at_ifr88.cnrs-mrs.fr>
> wrote:
>>
>> On 19 January 2011 13:53, Ajasja Ljubetič <ajasja.ljubetic_at_gmail.com>
>> wrote:
>> > <write histogram or have a running average of a colvar not in abf>
>> > I'm happy to report that  having multiple histogram blocks
>> > in addition to
>> > the ABF block works beautifully. At the same time it's also possible to
>> > run
>> > the analysis.
>> > One question regarding the runAveStride:
>> > Is this the number of steps between two points inside the length window?
>> > so
>> > if I have runAveStride 10 and runAveLength 100, then it will only
>> > average
>> > ten points?
>> > What I would really like is to output every 100th step the average for
>> > the
>> > last 100 steps.
>>
>> This is not strictly speaking a running average, and not supported by
>> the current code. Your best choice here is to save a colvar trajectory
>> with sufficient time resolution.
>> Cheers,
>> Jerome
>
>

This archive was generated by hypermail 2.1.6 : Wed Feb 29 2012 - 15:56:33 CST