Hi HM,
I would take a good look in the literature at what people do in these
situations. The issue with your 'control' is that you make no mention of
timescale. The experimental data may indeed show that the peptide
penetrates the membrane but does it give any indication of the timescale
that this takes? If it is on the order of milliseconds then just running
brute force MD is not going to work. One option would be to try some form
of umbrella sampling where you effectively move the peptide through the
membrane in stages. This will give you a PMF for crossing the membrane.
While non membrane interacting peptides will also move into the membrane
in this situation the difference is that any barriers you observe in the
PMF should be much higher. The same is true if you run steered MD,
although it will not be quantitative, the behavior will depend on if you
do constant velocity or constant force. The issue is that entropy is
likely to be against you since there are a large number of orientational
configurations for the peptide in solution but likely only one major
orientation within the membrane. This is probably part of the issue why
you see diffusion away from the membrane in the unbiased simulation.
Good luck.
All the best
Ross
On 1/18/13 12:28 AM, "HM" <scienceamber.gmail.com> wrote:
>Yeah I thought of steered molecular dynamics at first glance. But then I
>am
>afraid because in that case, it will forcefully pull the peptide through
>the lipid. So I will get the penetration behavior for non-penetrating
>peptide as well. :-(
>
>HM
>
>On Fri, Jan 18, 2013 at 9:22 AM, Thomas Evangelidis
><tevang3.gmail.com>wrote:
>
>> Good question. I think you can deactivate the restraint when you see the
>> peptide to enter the bilayer. Alternatively, in the flat-well restraint
>>you
>> define 4 cutoff distances (r1,r2,r3,4) (have a look in older threads).
>>if
>> the distance between the peptide and the lipid headgroup (R) is r2<R<r3
>> then no force is applied. Thereby you can use values for those 2 cutoffs
>> such that r3-r2>=[length of bilayer]. But I think the former solution
>>would
>> be better.
>>
>> Also you may want to consider steered MD, this is what people use in
>>cases
>> like yours.
>>
>> Thomas
>>
>>
>> On 18 January 2013 10:10, HM <scienceamber.gmail.com> wrote:
>>
>> > Thanks Thomas for your reply.
>> > Just one quick query. If I will keep the flat-well restrain on peptide
>> > close to lipid head group. Then I expect that there will be no moving
>> away
>> > phenomenon but at the same time is it not true that it will also not
>>go
>> > inside the membrane due to restrain w.r.t. lipid head group ?
>> >
>> > Thanks,
>> > HM
>> >
>> > On Fri, Jan 18, 2013 at 8:52 AM, Thomas Evangelidis <tevang3.gmail.com
>> > >wrote:
>> >
>> > > You can apply a flat-well restraint that will keep the peptide close
>> to a
>> > > lipid head group and then just pray for your peptide to pass through
>> the
>> > > bilayer. I would certainly use an enhanced sampling algorithm to
>> monitor
>> > > this phenomenon, but as far as I know membranes are peculiar systems
>> and
>> > > the conventional aMD algorithm implemented in AMBER is not suitable
>>for
>> > > that type of simulation. I would search the literature to see what
>> other
>> > > people have used for similar cases.
>> > >
>> > > Thomas
>> > >
>> > >
>> > > On 18 January 2013 09:29, HM <scienceamber.gmail.com> wrote:
>> > >
>> > > > Hi Ben,
>> > > >
>> > > > *"You placed the peptide next to the lipid bilayer with VMD. It
>> sounds
>> > > like
>> > > > you expect the peptide insert into the membrane and cross the
>> bilayer,
>> > > but
>> > > > it diffused away. "*
>> > > > Yes this is what exactly I did, I expect and I am getting. :-(
>> > > >
>> > > > *"you'll probably need a large amount of sampling in hopes of
>>seeing
>> a
>> > > > peptide-bilayer interaction. "*
>> > > > If I understand you correctly, Shall I continue previous
>>simulation
>> and
>> > > > hope for peptide-bilayer interaction
>> > > >
>> > > > *Why is this a control? What are you measuring or investigating in
>> this
>> > > > study?*
>> > > > This is control because It is already proven (in several
>>experiments)
>> > > that
>> > > > given peptide interacts with membrane and finally crosses it. And
>>I
>> > want
>> > > to
>> > > > keep it as control because I have bunch of peptides in my hand,
>>for
>> > > which I
>> > > > am interested in testing their membrane crossing behavior. The one
>> > which
>> > > > will give positive result, I will use further for my lab work.
>> > > >
>> > > > *Is this really an appropriate method/approach? *
>> > > > I don't know whether it is an appropriate approach or not. I
>>thought
>> > > about
>> > > > it and trying to perform the simulations. If you can correct me or
>> > > suggest
>> > > > something then please do it. Just for your info, I have amber10
>>and
>> > > amber11
>> > > > and not amber12 at my system. So for all my work, I can use only
>>both
>> > of
>> > > > them and not Amber12 (which recently added membrane force-fields
>>with
>> > > nice
>> > > > tutorial).
>> > > >
>> > > > Thanks,
>> > > > HM
>> > > >
>> > > >
>> > > >
>> > > _______________________________________________
>> > > 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
>> >
>>
>>
>>
>> --
>>
>> ======================================================================
>>
>> Thomas Evangelidis
>>
>> PhD student
>> University of Athens
>> Faculty of Pharmacy
>> Department of Pharmaceutical Chemistry
>> Panepistimioupoli-Zografou
>> 157 71 Athens
>> GREECE
>>
>> email: tevang.pharm.uoa.gr
>>
>> tevang3.gmail.com
>>
>>
>> website: https://sites.google.com/site/thomasevangelidishomepage/
>> _______________________________________________
>> 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 Fri Jan 18 2013 - 01:00:03 PST
