[gmx-users] Issues combining protein and membrane file?

[Justin Lemkul]

On 2/18/17 4:46 PM, Sanim Rahman wrote:
> Hi all,
>
> I am trying to prepare a system for a MD simulation of an ion channel
> (Kv1.2) in a POPC lipid bilayer. I am struggling inserting the protein into
> the lipid membrane. I am currently trying the inflategro.pl approach done
> in the KALP15 tutorial. The first step with the prepared structural files
> was to combine them together using
>
> cat protein.gro lipid.gro > system.gro
>
> However, after combining them and loading them into VMD, my system has
> become distorted and only 151 atoms out 90131 atoms were read. Is there
> anything I can do to fix this issue?
>
> I believe it may have to do with the nomenclature of my script. The .gro
> file for my protein and lipid bilayer is not written in a consistent format:
>
>   421THR     CA10936   4.791   7.926   7.910
>   421THR     CB10937   4.822   7.797   7.831
>   421THR    OG110938   4.862   7.831   7.698
>   421THR    HG110939   4.882   7.748   7.647
>   421THR    CG210940   4.701   7.705   7.826
>   421THR      C10941   4.754   7.892   8.055
>   421THR     O110942   4.772   7.974   8.146
>   421THR     O210943   4.649   7.808   8.074
>     1POPC    C1    1   5.168   4.220   4.016 -0.2491  0.0715  0.4836
>     1POPC    C2    2   4.942   4.276   3.936  0.1967  0.1087 -0.5850
>     1POPC    C3    3   5.108   4.444   3.993  0.3211  0.0553 -0.2150
>     1POPC    N4    4   5.085   4.311   3.935  0.2251  0.0126 -0.0777
>     1POPC    C5    5   5.129   4.301   3.795 -0.0894 -0.4945 -0.1460
>     1POPC    C6    6   5.280   4.304   3.767  0.0019 -0.2403  0.3568
>     1POPC   OS7    7   5.354   4.422   3.799  0.0579 -0.0618 -0.4147
>     1POPC    P8    8   5.512   4.403   3.778  0.1576  0.2198  0.0575
>     1POPC   OM9    9   5.532   4.289   3.686  0.1757 -0.1049  0.4662
>     1POPC  OM10   10   5.597   4.524   3.785  0.1671  0.2024  0.2449
>     1POPC  OS11   11   5.526   4.331   3.921  0.4792 -0.0093 -0.0879
>

VMD is probably choking on the fact that some of your lines have velocity 
information (the lipids) and others (the protein) don't.  Try stripping the 
velocities and appending just the coordinates.

> I updated the number of atoms in my system and removed nonessential lines
> as well. Also, would the inflategro procedure be appropriate for this
> simulation considering that I am inserting a +10,000 atom protein into a
> lipid membrane?
>

Sure, we've done it with big systems, though the script is a bit of a memory 
hog, IIRC.  Haven't used it in a while.

-Justin

-- 
==================================================

Justin A. Lemkul, Ph.D.
Ruth L. Kirschstein NRSA Postdoctoral Fellow

Department of Pharmaceutical Sciences
School of Pharmacy
Health Sciences Facility II, Room 629
University of Maryland, Baltimore
20 Penn St.
Baltimore, MD 21201

jalemkul at outerbanks.umaryland.edu | (410) 706-7441
http://mackerell.umaryland.edu/~jalemkul

==================================================