From: Jie Shi (shijie0204_at_tamu.edu)
Date: Tue Sep 13 2022 - 02:03:35 CDT
Hi Norman,
Thank you for your explanations! I will try this and hopefully, it will
solve the problem (fingers crossed!)
Thanks again!
Best,
Jie
On Mon, Sep 12, 2022 at 11:40 PM Geist, Norman <
norman.geist_at_uni-greifswald.de> wrote:
> Using "useconstantarea" will cause the box to only change in the
> Z-dimension, useful to only equilibrate the solvent above and below a
> surface with fixed bulk properties. Using "useconstantratio" will keep the
> ratio between X and Y constant,
> ZjQcmQRYFpfptBannerStart
> This Message Is From an External Sender
> This message came from outside your organization.
>
> ZjQcmQRYFpfptBannerEnd
> Using "useconstantarea" will cause the box to only change in the
> Z-dimension, useful to only equilibrate the solvent above and below a
> surface with fixed bulk properties. Using "useconstantratio" will keep the
> ratio between X and Y constant, also useful wich e.g. a crystal surface. In
> your case, you may just plain want "useflexiblecell" with none of the
> above, so each dimension can relax on its own.
>
> Bests
> Norman Geist
>
> Am Dienstag, den 13-09-2022 um 07:36 schrieb Jie Shi:
>
> I tried "useflexiblecell" along with "useConstantArea" (replacing the
> constant ratio). But no, I have not tried just using "useflexiblecell"
> without "useConstantArea" or "useconstantratio".
>
> This idea is new to me. Do you think this may be the cause?
>
> Best,
> Jie
>
>
> On Mon, Sep 12, 2022 at 10:31 PM Geist, Norman <
> norman.geist_at_uni-greifswald.de> wrote:
>
>> Have you tried without useconstantratio? Am Dienstag, den 13-09-2022 um
>> 07: 28 schrieb Jie Shi: Hi Norman, Thanks for the clarification. Yes, I did
>> have both "useflexiblecell" and "useconstantratio" set on in my input
>> configuration files, but
>> ZjQcmQRYFpfptBannerStart
>> This Message Is From an External Sender
>> This message came from outside your organization.
>>
>> ZjQcmQRYFpfptBannerEnd
>> Have you tried without useconstantratio?
>>
>>
>> Am Dienstag, den 13-09-2022 um 07:28 schrieb Jie Shi:
>>
>> Hi Norman,
>>
>> Thanks for the clarification. Yes, I did have both "useflexiblecell" and
>> "useconstantratio" set on in my input configuration files, but still, I got
>> no luck. The membrane bending is still there...
>>
>> Best,
>> Jie
>>
>>
>>
>>
>>
>> On Mon, Sep 12, 2022 at 10:22 PM Geist, Norman <
>> norman.geist_at_uni-greifswald.de> wrote:
>>
>>> I'd go with an NPT equilibration step that includes "useflexiblecell" to
>>> allow each dimension to relax independently. The lipids just ask for more
>>> room in one dimension, which cannot be resolved with "useconstantratio".
>>> Bests
>>> ZjQcmQRYFpfptBannerStart
>>> This Message Is From an External Sender
>>> This message came from outside your organization.
>>>
>>> ZjQcmQRYFpfptBannerEnd
>>> I'd go with an NPT equilibration step that includes "useflexiblecell" to
>>> allow each dimension to relax independently. The lipids just ask for more
>>> room in one dimension, which cannot be resolved with "useconstantratio".
>>>
>>> Bests
>>> Norman Geist
>>>
>>> Am Dienstag, den 13-09-2022 um 00:17 schrieb Jie Shi:
>>>
>>> Hi Raman,
>>>
>>> Thank you for your reply. The snapshot showing the membrane bending can
>>> be found here: https://urldefense.com/v3/__https://postimg.cc/jWrpgPFH__;!!DZ3fjg!4_m_XAV9fqCM3YmU557Wy4vdUVz3ZaCFC5wB_iQigrkNjvZPQ1Abk9pvWEmM8YIQ36jvoAib7Kv_xsvtoPWIpOA$
>>> <https://urldefense.com/v3/__https://postimg.cc/jWrpgPFH__;!!DZ3fjg!8SGM2Y0Iiu5eCxBnsVlOZz8ct9MyjB_EKDU41jkwZp3ebOuLMORPoAO47xBkkAO36ib1bihdLbpiEmoIszuz18cuB8iBjg$>
>>>
>>>
>>> 1. Large membrane patches (>5-10 nm side) show membrane undulations.
>>> These are ripple-like movements/bending. Is the curvature you are looking
>>> at static or moves up-down/back-forth?
>>>
>>> Once the curvature is formed, it became quite stable and does not move
>>> much. It’s as if the bent form is the stabilized form for the membrane
>>> structure.
>>>
>>>
>>> 1. If these are undulations, you easily suppress these by
>>> restraining some of the atoms (e.g. phosphate groups of a few lipid
>>> molecules). Just apply restraints in the z-direction so the bulges don’t
>>> form but lipids can diffuse within the membrane (x-y direction).
>>>
>>> Yes, but our ultimate goal of the simulation is to study how the protein
>>> would affect the membrane, especially the potential capability of
>>> membrane-bending. For that purpose, we can not apply restraints as it will
>>> also prohibit the membrane change we would like to study.
>>>
>>>
>>> 1. In membranes, it is common to use semi-isotropic pressure
>>> coupling. Are you using isotropic coupling?
>>>
>>> We use constant P and T along with “useConstantRatio yes”, is this the
>>> semi-isotropic pressure coupling that you were referring to?
>>>
>>> Best,
>>> Jie
>>>
>>> On Mon, Sep 12, 2022 at 2:11 AM Raman Preet Singh <
>>> ramanpreetsingh_at_hotmail.com> wrote:
>>>
>>>> Hi Jie, I am unable to download the attachment so there's a bit of
>>>> guess work here. 1. Large membrane patches (>5-10 nm side) show membrane
>>>> undulations. These are ripple like movements/bending. Is the curvature you
>>>> are looking at is static
>>>> ZjQcmQRYFpfptBannerStart
>>>> This Message Is From an External Sender
>>>> This message came from outside your organization.
>>>>
>>>> ZjQcmQRYFpfptBannerEnd
>>>>
>>>> Hi Jie,
>>>> I am unable to download the attachment so there's a bit of guess work here.
>>>> 1. Large membrane patches (>5-10 nm side) show membrane undulations. These are ripple like movements/bending. Is the curvature you are looking at is static or moves up-down/back-forth?
>>>> 2. If these are undulations, you easily suppress these by restraining some of the atoms (e.g. phosphate groups of a few lipid molecules). Just apply restraints in the z-direction so the bulges don't form but lipids can diffuse within the membrane (x-y direction).
>>>> 3. In membranes, it is common to use semi-isotropic pressure coupling. Are you using isotropic coupling?
>>>> For membranes, i have worked on Gromacs but some of these principles are applicable to all MD programs.
>>>> Regards,
>>>> Raman
>>>> ________________________________
>>>> From: owner-namd-l-digest_at_ks.uiuc.edu <owner-namd-l-digest_at_ks.uiuc.edu> on behalf of namd-l digest <owner-namd-l-digest_at_ks.uiuc.edu>
>>>> Sent: Monday, September 12, 2022 1:06:19 PM
>>>> To: namd-l-digest_at_ks.uiuc.edu <namd-l-digest_at_ks.uiuc.edu>
>>>> Subject: namd-l digest V1 #3281
>>>> namd-l digest Monday, September 12 2022 Volume 01 : Number 3281
>>>> In this issue:
>>>> Re: namd-l: Fwd: Irregular spontaneous membrane bending even in the absence of protein
>>>> Re: namd-l: Fwd: Irregular spontaneous membrane bending even in the absence of protein
>>>> Re: namd-l: QwikMD QM Tut - FATAL ERROR: Error running command for QM forces calculation.
>>>> ----------------------------------------------------------------------
>>>> Date: Sun, 11 Sep 2022 15:38:29 -0500
>>>> From: Jeff Comer <jeffcomer_at_gmail.com>
>>>> Subject: Re: namd-l: Fwd: Irregular spontaneous membrane bending even in the absence of protein
>>>> It looks to me like you are doing everything right. Ib ve run a lot of
>>>> membrane simulations based on CHARMM-GUI structures and have never seen
>>>> anything similar. How long does it take the curvature to appear? Have you
>>>> tried running longer to see if the curvature disappears?
>>>> Jeff
>>>> On Sun, Sep 11, 2022 at 10:42 Jie Shi <shijie0204_at_tamu.edu> wrote:
>>>> > Dear all,
>>>> >
>>>> > I am attempting to run a simulation of a system containing both membrane
>>>> > (POPC) and protein with NAMD 2.14 on HPC. However, I found the membrane
>>>> > forms an irregular spontaneous bending. So I then tested it with a
>>>> > membrane-only system (without the protein, so only the POPC bilayer, water,
>>>> > and ions are included), and a similar bending is again shown up, which
>>>> > suggests itb s not the protein that induces the bending in the membrane
>>>>
>>>>
This archive was generated by hypermail 2.1.6 : Tue Dec 13 2022 - 14:32:44 CST