From: Richard Wood (rwoodphd_at_yahoo.com)
Date: Fri Sep 07 2007 - 07:20:40 CDT
Yes, but I am sure that physiological folding or unfolding of proteins didn't occur at 500 K...
Richard
Richard L. Wood, Ph. D.
University of Minnesota
Dept. of Medicinal Chemistry,
College of Pharmacy
717 Delaware St. SE
Minneapolis, MN
55414-2959
rwoodphd_at_yahoo.com
----- Original Message ----
From: "gs_at_mrc-lmb.cam.ac.uk" <gs_at_mrc-lmb.cam.ac.uk>
To: Arun Krishnan <krishnan_at_ttck.keio.ac.jp>
Cc: Richard Wood <rwoodphd_at_yahoo.com>; namd-l_at_ks.uiuc.edu
Sent: Friday, September 7, 2007 1:42:34 AM
Subject: Re: namd-l: Temperature dependent unfolding..
While Richard's remark applies to the general case, folding proteins
represent a special case for the way they evolved to get a unique native
conformation. Experimentlists do very nasty things to proteins (mutations,
truncations, addition of huge dyes, dramatic changes of solution
conditions) and the native conformation of the protein is only slightly
affected. Folding pathways, especially in the region close to the native
state, are a robust feature of the protein, as well. This, along with the
low resolution of available experimental data, are probably the reasons
why many MD unfolding studies (usually involving unphysical temperatures
and pressures) have found a substantial agreement with experimental data
on the transition state for folding. Even if the force field is not
optimally tuned at those conditions, the unfolding pathways are probably
only partially affected.
Anyway, great care should always be used when analysing these data, in
order to distinguish real protein features from force fields artifacts.
In a mechanical unfolding, you are selecting a very special pathway out of
those chaotropically available, that is why the two will not generally
coincide (unless the specific protein has evolved under the selective
pressure of mechanical unfolding).
Gianni
> Thanks to all for your suggestions/comments. I am a little confused
> though.
> Since the Phys Rev. paper claimed to see significant unfolding (> 12
> Angstroms) for simulations at 450 and 520K, they either used a higher
> pressure or something else. Does this mean that unfolding simulations, in
> general, are not to be trusted? I have seen papers where there is pretty
> good fit between experimental and simulated conformations for the
> transition
> state etc. Just curious to know what kind of results others have obtained
> here.
>
> Also, in a paper which compares thermal unfolding with mechanical
> unfolding
> (such as is given in the NAMD tutorial), there seem to be significant
> differences in the unfolding pathway for the two methods..
>
> All very confusing!!
>
> Cheers,
>
> Arun
>
> On 9/7/07, Richard Wood <rwoodphd_at_yahoo.com> wrote:
>>
>> As I pointed out to Giovanni earlier, as much as we'd like to be able
>> to,
>> sometimes we can't simulate accurately such processes, as sometimes
>> there is
>> a limit as to how we can model the physics of certain situations (i.e,
>> the
>> diffusion of lipids in a membrane, or the flow of water over a crystal
>> surface of cellulose, two things I've simulated that wound up having
>> "supersonic" velocities), due to the limitations of the code.
>>
>> I personally wouldn't put too much stock in someone trying to compute
>> the
>> unfolding of proteins at this point in time.
>>
>> Richard
>>
>> Richard L. Wood, Ph. D.
>> University of Minnesota
>> Dept. of Medicinal Chemistry,
>> College of Pharmacy
>> 717 Delaware St. SE
>> Minneapolis, MN 55414-2959
>> rwoodphd_at_yahoo.com<https://mail.google.com/mail?view=cm&tf=0&to=rwoodphd@yahoo.com>
>>
>>
>>
>> ----- Original Message ----
>> From: Neema Salimi
>> <nsalimi_at_msg.ucsf.edu<https://mail.google.com/mail?view=cm&tf=0&to=nsalimi@msg.ucsf.edu>
>> >
>> To: Giovanni Settanni
>> <gs_at_mrc-lmb.cam.ac.uk<https://mail.google.com/mail?view=cm&tf=0&to=gs@mrc-lmb.cam.ac.uk>
>> >
>> Cc:
>> namd-l_at_ks.uiuc.edu<https://mail.google.com/mail?view=cm&tf=0&to=namd-l@ks.uiuc.edu>
>> Sent: Thursday, September 6, 2007 12:45:22 PM
>> Subject: Re: namd-l: Temperature dependent unfolding..
>>
>> Extrapolations from room temperature (or slightly higher) are pretty
>> worthless for simulations at these temperatures. I've had proteins
>> bigger than ubiquitin that unfold far slower at room temperature that
>> unfold just fine at 500K. And extrapolations based on (delta)Cp are
>> equally meaningless, as for some of the proteins I've looked at, says
>> they would unfold in less than 1 picosecond, obviously quite impossible.
>>
>> On Sep 6, 2007, at 4:34 AM, Giovanni Settanni wrote:
>>
>> > Hi Arun,
>> >
>> > did you (very very roughly) check the expected unfolding time of
>> > ubiquitin at high temperature extrapolating the Arrhenius's law
>> > (with the brutal assumption that there are no solvent phase
>> > transitions as you would not see them in the simulations)? It may
>> > be much larger than
>> > 12ns, and your simulations may be correct (even if they provide no
>> > information about the unfolding pathways!). Ubiquitin has a very small
>> > unfolding rate at 300K (0.001 s-1), I do not expect it to increase by
>> > almost 12 orders of magnitudes with temperature alone. The authors
>> > of the phys rev E probably did a temperature AND pressure induced
>> > unfolding.
>> > Cheers
>> >
>> > Gianni
>> >
>> > --
>> > Giovanni Settanni
>> > Centre for Protein Engineering
>> > MRC Centre
>> > Hills Road,
>> > Cambridge, UK
>> > CB2 0QH
>> > Phone: +44 1223 402133
>> > Fax: +44 1223 402140
>> >
>> >
>> > Arun Krishnan wrote:
>> >> Hi Monika,
>> >> I have uploaded all my config files to my website. You can get
>> >> them from here:
>> >> a) energy minimization:
>> http://www.iab.keio.ac.jp/~krishnan/
This archive was generated by hypermail 2.1.6
: Wed Feb 29 2012 - 15:45:13 CST