Re: Bad structure after converting the coarse grained last frame to all-atom

From: Rabeta Yeasmin (rabetayeasmin_at_gmail.com)
Date: Thu Sep 20 2018 - 09:58:39 CDT

The minimization shows a lot of warning like-

Warning: Bad global bond count! (74224 vs 74704)
Warning: Bad global angle count! (88963 vs 91517)
Warning: Bad global dihedral count! (96371 vs 102836)
Warning: Low global exclusion count! (251495 vs 268840)

Warning: Tuple 38411 with atoms 14385(180) 14384(180) 14396(184) 14398(184)
missing patch 184

And the stretching does not remove even affter minimization.

Rabeta Yeasmin

On Wed, Sep 19, 2018 at 8:52 PM Peter Freddolino <petefred_at_umich.edu> wrote:

> Hm, not sure then what would be causing this. It is possible that you're
> just seeing the best that can be done given the cg conformation. Does it
> end up minimizing ok?
>
> On Wed, Sep 19, 2018 at 2:00 PM, Rabeta Yeasmin <rabetayeasmin_at_gmail.com>
> wrote:
>
>> Dear Mr Peter,
>>
>> I have used text interface to reverse the structure using the value 1 for
>> do_rotations argument. but the structure looks still stretched out.
>>
>> Rabeta Yeasmin
>>
>>
>> On Tue, Sep 18, 2018 at 1:36 PM Rabeta Yeasmin <rabetayeasmin_at_gmail.com>
>> wrote:
>>
>>> Thanks for your suggestion. I am just reconverting the proteins and
>>> lipids. I will try the way you suggested soon and let you know what I find.
>>>
>>> Rabeta Yeasmin
>>>
>>> On Tue, Sep 18, 2018, 12:27 PM Peter Freddolino <petefred_at_umich.edu>
>>> wrote:
>>>
>>>> It looks like the do_rotations argument is set by default in the gui,
>>>> but just to confirm, could you try doing the reversal via the text
>>>> interface (see https://www.ks.uiuc.edu/Research/vmd/plugins/cgtools/)
>>>> for directions). Be sure to specify the value 1 for the do_rotations
>>>> argument, which should do its best to minimize the perturbations to the
>>>> reversed structure. Also, are you trying to reverse cg the water and ions?
>>>> I would not recommend doing so, but rather, reversing the lipid/protein and
>>>> then re-solvating.
>>>> Best,
>>>> Peter
>>>>
>>>> On Tue, Sep 18, 2018 at 11:58 AM, Rabeta Yeasmin <
>>>> rabetayeasmin_at_gmail.com> wrote:
>>>>
>>>>> Dear Mr. Peter,
>>>>>
>>>>> I have used CG Builder GUI in VMD to reverse the coarse-grained
>>>>> molecule. That did not ask me to set do_rotations argument. And my system
>>>>> contains two protein, POPC and POPG lipid, water and ions. I have used the
>>>>> RBCG model that is provided in VMD 1.9.2 as plugins. I think it is the old
>>>>> RBCG model from 2007.
>>>>>
>>>>> Thanks.
>>>>>
>>>>> Rabeta Yeasmin
>>>>>
>>>>> On Mon, Sep 17, 2018 at 3:58 PM Peter Freddolino <petefred_at_umich.edu>
>>>>> wrote:
>>>>>
>>>>>> Can you please clarify the precise command that you used to reverse
>>>>>> coarse grain the molecule? Did you have the do_rotations argument set? What
>>>>>> does your system contain?
>>>>>> Also, what CG model are you using? I would hope a recent martini
>>>>>> version and not the old RBCG model from 2007, unless you're trying to
>>>>>> precisely reproduce one of those old papers...
>>>>>> Best,
>>>>>> Peter
>>>>>>
>>>>>> On Mon, Sep 17, 2018 at 12:50 PM, Rabeta Yeasmin <
>>>>>> rabetayeasmin_at_gmail.com> wrote:
>>>>>>
>>>>>>> Dear NAMD users,
>>>>>>>
>>>>>>> I have run some coarse-grained simulation of a protein-lipid system
>>>>>>> in NAMD using residue-based coarse-grained simulation. After running 500ns,
>>>>>>> I have converted back the last structure of coarse-grained simulation to
>>>>>>> the all-atom system. But the structure has a lot of stretching. And during
>>>>>>> the minimization of the converted structure, it showed a lot f warning like-
>>>>>>> Warning: Tuple 540098 with atoms 20552(159) 20548(159) missing patch
>>>>>>> 129
>>>>>>>
>>>>>>> I have tried another way to overcome this kind of problem. I have
>>>>>>> minimized the last structure of coarse-grained simulation before
>>>>>>> converting it to the all-atom system. But that did not work too. Can you
>>>>>>> please suggest me what can cause this kind of problem?
>>>>>>> Thanks.
>>>>>>>
>>>>>>> Rabeta Yeasmin
>>>>>>>
>>>>>>
>>>>>>
>>>>
>

This archive was generated by hypermail 2.1.6 : Sun Dec 08 2019 - 23:20:09 CST