Re: Protein Charge

From: Brian Radak (brian.radak_at_gmail.com)
Date: Mon Jan 07 2019 - 10:39:28 CST

Generally speaking, there's no such strict requirement for the system to be
neutral (although this is best practice). When using straightforward PME
the forces are well-defined no matter the net charge (this was argued
somewhat heavily in the literature about 20 years ago, see the work of Tom
Cheatham, for example). The energy, however, is somewhat ill-defined when
the box is not neutral, but this does not matter much unless you are making
direct energy comparisons between multiple systems and/or boxes. The effect
generally vanishes with larger boxes, but can persist quite seriously in
certain cases (most notably, membrane systems). Adding an electric field might
change some of these things, but I'm not up to date on that literature.

Also, it should be noted that PME also causes rather unphysical rotation
and translation of charged biomolecules, especially when the charge is
large (e.g. nucleic acids) and the box is small and/or highly anisotropic
(e.g. an orthorhombic cell with a single long axis). This is because the
system can artificially lower its energy by aligning the molecular dipole
with the shorter axis, since then it is closer to its own image. This is a
well-known problem for large RNA and DNA systems, where the problem is
often solved by extra orientational restraints (which can be expensive!)
and the bias is assumed to be less bad than the artifact.

HTH,
BKR

On Mon, Jan 7, 2019 at 11:19 AM Peter Freddolino <petefred_at_umich.edu> wrote:

> Just to be clear, the *system* as a whole should be neutral, but the
> protein itself need not be (and frequently won't be). So one can very
> easily have a situation where charged macromolecules will move in response
> to an applied voltage, but the system as a whole is still net neutral.
> Best,
> Peter
>
> On Mon, Jan 7, 2019 at 8:45 AM Ashkan Shekaari <shekaari_at_email.kntu.ac.ir>
> wrote:
>
>> Thank you so much, dear Soroush.
>>
>> --
>> Regards,
>> Ashkan Shekaari,
>> Ph.D Candidate in Solid State Physics,
>> K. N. Toosi University of Technology,
>> Tehran, Iran
>>
>> ------------------------------
>> *From: *"soroush ziaei" <soroush.ziaei7337_at_gmail.com>
>> *To: *"namd-l" <namd-l_at_ks.uiuc.edu>, "shekaari" <
>> shekaari_at_email.kntu.ac.ir>
>> *Sent: *Monday, January 7, 2019 5:07:06 PM
>> *Subject: *RE: namd-l: Protein Charge
>>
>> Yes, the sum of charges of your structures should be zero, it is
>> necessary.
>>
>> By the way, you can add some ion to structure in order to achieve a
>> neutral charge.
>>
>>
>>
>> Good luck, ashkan
>>
>>
>>
>> Sent from Mail
>> <https://urldefense.proofpoint.com/v2/url?u=https-3A__go.microsoft.com_fwlink_-3FLinkId-3D550986&d=DwMFaQ&c=OCIEmEwdEq_aNlsP4fF3gFqSN-E3mlr2t9JcDdfOZag&r=jUfnSyKkfkyVRBIUzlG1GSGGZAZGcznwr8YliSSCjPc&m=SxO23DkirmDEhBY_RtMrHLmtS-RVK1P7kWKIfa11i7k&s=rliZ6UKMLqoA0mxAvHqCE9M1AhE8BTKJbCUIBVZiNPg&e=>
>> for Windows 10
>>
>>
>>
>> *From: *Ashkan Shekaari <shekaari_at_email.kntu.ac.ir>
>> *Sent: *Monday, January 7, 2019 11:56 AM
>> *To: *namd-l <namd-l_at_ks.uiuc.edu>
>> *Subject: *namd-l: Protein Charge
>>
>>
>>
>> *Dear Users,*
>>
>>
>>
>> *When I simulate my protein in the NVT ensemble and apply an electric
>> field, *
>>
>> *the protein begins to move along the direction of the field.* *Which
>> means that *
>>
>> *the protein is not charge-neutral.*
>>
>> *Is my simulation accordingly wrong?*
>>
>> *Is it necessary for a protein to be charge-neutral in MD simulations?*
>>
>>
>>
>> *--*
>> *Regards,*
>> *Ashkan Shekaari,*
>> *Ph.D Candidate in Solid State Physics,*
>> *K. N. Toosi University of Technology,*
>> *Tehran, Iran*
>>
>>
>> --
>> This message has been scanned for viruses and dangerous content by
>> *KNTU Antispam System (E.F.A. Project)*
>> <https://urldefense.proofpoint.com/v2/url?u=http-3A__www.efa-2Dproject.org&d=DwMFaQ&c=OCIEmEwdEq_aNlsP4fF3gFqSN-E3mlr2t9JcDdfOZag&r=jUfnSyKkfkyVRBIUzlG1GSGGZAZGcznwr8YliSSCjPc&m=ILzoZkBQwl1NU0ot9TlDgtr2uJrObVxgbHaI2K9LbTw&s=ycJxk2lepzyMOiRhsIxpqYBRxJNBZ66pf0F_y-oEipA&e=>,
>> and is believed to be clean.
>>
>>
>>
>>

This archive was generated by hypermail 2.1.6 : Tue Dec 10 2019 - 23:20:23 CST