[gmx-users] Can't unfold the protein

[felmerino at uchile.cl]

Hey,

 

NPT is not the appropriate way to do this kind of simulations. I am not sure whether or not the water models available for classic MD simulations are able to reproduce the phase behavior. Indeed what you see when your system explodes and gets huge is the water evaporating. What is generally done is to allow the system to reach the density of liquid water at the temperature that you desire (see the Dagget's papers of that) and then use NVT for the rest of the simulation. 

Regarding the kinetics of the process it is up to the protein. There are some proteins that unfold really fast and others reallly really slow. For instance, on one of my systems i have to simulate 50 ns at 600 K to only see a 40% decrease in structure. 

As you may already imagine, the pressure gets quite high in these kind of simulations. Extreme pressure can also induce protein unfolding, but the depence here is again protein dependent since moderately high pressure tends to stabilize the structure. Of course what is moderate and what is high is case dependent.

Please consider also that biomolecular forcefields were not designed with these extreme temperatures in mind, so while doing high temperature induced unfolding you are taking the FF to the limit and you should be extremely careful of what kind of conclusion you obtain from your simulation.

 

Regards

 

Felipe
----Mensaje original----De: chris.neale at utoronto.caFecha: 16-jun-2011 12:58Para: <gmx-users at gromacs.org>Asunto: [gmx-users] Can&amp;#39;t unfold the proteinDear Hsin-Lin:I am no expert in this area. I am just saying that if you get  densities that are way too low in NPT, then you might alleviate this  problem with NVT.In fact, I would personally do this in vacuum without pbc and use the  sd integrator. Then you can sample extended conformations. I'd  consider this a method to generate unfolded conformations that  probably have no relation to the true temperature denatured state. But  let's be honest, you're never going to come close to Boltzmann  sampling of an unfolded state in 300 ns so why bother with the water?Your link ( http://manual.gromacs.org/online/protunf.html ) is only  concerned with analysis, so it is irrelevant.If you are determined to use NPT then I suppose that you could use the  Berendsen barostat (more stable in simulations but you get the wrong  ensemble) with a very low compressibility. That's about all I can say,  perhaps somebody else will comment.Good luck,Chris.-- original message --Dear Chris,Thank you for your reply.My protein is very stable.I simulated it in 300ns in 300K before but there was almost no change.That's why I want to do denaturation now.I want to start a new simulation on the protein which is unfolded.And I'll read the paper you told me.Thank you.But I don't understand what you recommend me to do.You use 3000K on your protein and the protein unfolded.I use 600K on my protein but the system explode(box become very big  and protein locate at corner).I saw the second way on this web,http://manual.gromacs.org/online/protunf.htmlBut it also useless to me.And in my mdp I use thermostat and barostat, which means my system is NPT.Does anything wrong in my methods?Sincerely yours,Hsin-Lin--gmx-users mailing list    gmx-users at gromacs.orghttp://lists.gromacs.org/mailman/listinfo/gmx-usersPlease search the archive at http://www.gromacs.org/Support/Mailing_Lists/Search before posting!Please don't post (un)subscribe requests to the list. Use thewww interface or send it to gmx-users-request at gromacs.org.Can't post? Read http://www.gromacs.org/Support/Mailing_Lists

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://maillist.sys.kth.se/pipermail/gromacs.org_gmx-users/attachments/20110616/8532324e/attachment.html>