Menu

electron density and order parameters for AMBER lipid11 ff

2013-11-19
2018-03-27
  • Thomas Evangelidis

    Greetings,
    https://sourceforge.net/p/membplugin/discussion/create_topic/general/#
    I have simulated a POPE bilayer using the lipid11 ff and NAMD. The box I used was a hexagonal prism, so in order to calculate correct Area per Lipid I wrapped the coordinates of every frame to an orthorhombic box using the following tcl commands in VMD:

    pbc wrap -all -orthorhombic -compound fragment -center bb -centersel "resname OL and name C118"
    set frameNum 1000 ;# set the number of frames in the trajecotry
    for {set i 0} {$i < $frameNum} {incr i} {
    animate goto $i
    #pbc set [list [lreplace [string trim [pbc get] }{ ] 5 5 90.0]]
    molinfo top set gamma 90.0
    }
    pbc box -center origin -orthorhombic -center bb

    Then saved the trajectory into a new .dcd file and reopened it in VMD. I don't know if this procedure is correct but the Area per Lipid plot I got was in agreement with the experimental values (~62 A^2).

    I also tried to measure the SDC order parameters without success. Below is the output from membranescd along with the error message. One possible reason may be the modular nature of lipid11 residues, namely the lipid headgroup has resname PE and the two different alkyl chains PA and OL. Can you understand what the problem is?

    ######################### Tk Console Output

    loading history file ... 48 events added
    Main console display active (Tcl8.5.6 / Tk8.5.6)
    Membrane SCD: need atom selection
    Membrane SCD) Executing: membranescd -structure {/home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_hexbox.lipid11.prmtop} -trajectory {/home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_sample2frames.dcd} -lipid {PE PA OL} -leaf {0 1} -chain {2 3} -from {0} -to {1} -step {1} -sel {resname OL PA PE} -o {/home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_sample2frames.orderparSCD} -wrapcmd {}
    Membrane SCD) Calculating Order parameters...
    Building topology...
    known residues: OL PA PE
    residue: OL
    has no P headgroup, using name C12 as head instead
    assigning acyl chain id 1 according to C names
    index numbers appear to be correct
    has 1 acyl chains:
    chain id 1 with 17 carbon atoms
    residue: PA
    has no P headgroup, using name C116 as head instead
    assigning acyl chain id 1 according to C names
    index numbers appear to be correct
    has 1 acyl chains:
    chain id 1 with 15 carbon atoms
    residue: PE
    has no P headgroup, using name C11 as head instead

    ############################ ERROR MESSAGE

    Something went wrong.

    Command: membranescd -structure /home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_hexbox.lipid11.prmtop -trajectory /home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_sample2frames.dcd -lipid {PE PA OL} -leaf {0 1} -chain {2 3} -from 0 -to 1 -step 1 -sel {resname OL PA PE} -o /home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_sample2frames.orderparSCD -wrapcmd {}

    Error:
    can't read "ntopo": no such variable
    can't read "ntopo": no such variable
    while executing
    "return $ntopo"
    (procedure "fixupChainNumbers" line 23)
    invoked from within
    "fixupChainNumbers $ttopo"
    (procedure "::membranetool_common::topology::setOccupancyBetaFromTopolog..." line 39)
    invoked from within
    "::membranetool_common::topology::setOccupancyBetaFromTopology "$atomselection and resname $lipids" 10"
    (procedure "::membranescd::orderPar" line 23)
    invoked from within
    "::membranescd::orderPar"
    (procedure "::membranescd::scd" line 88)
    invoked from within
    "::membranescd::scd -structure /home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_hexbox.lipid11.prmtop -trajectory /home/thomas/Do..."
    ("eval" body line 1)
    invoked from within
    "eval ::membranescd::scd $args "
    invoked from within
    "if $errflag { error "Something went wrong.\n\nCommand: membranescd $args\n\nError: \n$errMsg" $savedInfo $savedCode }"
    (procedure "::membranescd::membranescd" line 16)
    invoked from within
    "::membranescd::membranescd -structure /home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_hexbox.lipid11.prmtop -trajectory /home/t..."
    ("eval" body line 1)
    invoked from within
    "eval ::membranescd::membranescd $args"
    (procedure "membranescd" line 1)
    invoked from within
    "membranescd -structure {/home/thomas/Documents/Transmembrane_Proteins/POPE_patch/prod/cMD/POPE_hexbox.lipid11.prmtop} -trajectory {/home/thomas/Docume..."
    ("eval" body line 1)
    invoked from within
    "eval membranescd $command_line"
    (procedure "::membranescdgui::scd" line 86)
    invoked from within
    "::membranescdgui::scd"
    invoked from within
    ".mtscdgui.buttons.runbutton invoke"
    ("uplevel" body line 1)
    invoked from within
    "uplevel #0 [list $w invoke]"
    (procedure "tk::ButtonUp" line 22)
    invoked from within
    "tk::ButtonUp .mtscdgui.buttons.runbutton"
    (command bound to event)

     
    • Ismael Rodriguez Espigares

      Try not including the head group resname (PE) in atom selection, and choosing only chain 2. The plugin will take chains as single molecules and will compute their SCD.

      If you want to visualize plots, compute only one resname each time or the different resnames in atom selection will be average in plot (but not in text output file).

       

      Last edit: Ismael Rodriguez Espigares 2013-11-20
      • Thomas Evangelidis

        I wish I have read this earlier. I wasted my whole day trying to fix atom names to make the plugin read both chains at the same time and at then end I got the average of both plots :(

         
  • Toni

    Toni - 2013-11-19

    I think AMBER11 breaks the assumption 1 residue - 1 lipid. Try to rename and renumber them in VMD to make them look as usual.

     

    Last edit: Toni 2013-11-19
  • Physics

    Physics - 2018-03-27

    Hi all,
    I am experiencing similar errors as Thomas:
    Membrane SCD) Calculating Order parameters...
    Building topology...
    known residues: SDPS SOPS
    residue: SDPS
    assigning acyl chain id 2 according to C names
    assigning acyl chain id 3 according to C names
    index numbers appear to be correct
    has 2 acyl chains:
    chain id 2 with 22 carbon atoms
    chain id 3 with 18 carbon atoms
    residue: SOPS
    has no P headgroup, using name N as head instead
    Error: Something went wrong.
    Error:
    can't read "ntopo": no such variable
    can't read "ntopo": no such variable
    ... etc ...

    I have tried as suggested to exclude the headgroup using the atom selection:
    not name N HN1 HN2 HN3 C13 O13A O13B C12 H12A C11 H11A H11B P O11 O12 O13 O14

    Or even just selecting the tail:
    name C31 C32 H2X H2Y ... etc ...

    However, this still results in the same errors as above. I'm not sure if this is the issue with my atom selection? Please kindly advise.
    Many thanks.

    PS. I am using a .gro and .xtc file in the CHARMM force field.

     

    Last edit: Physics 2018-03-27
    • Ismael Rodriguez Espigares

      SOPS seems to be the one giving trouble.
      Did you rename your system in a way that atoms of one phophoslipid molecule only has only 1 unique residue name and each atom has a unique atom name?

      If so, please, use de following atomselection: "resname SOPS", and copy paste the output of the VMD terminal/console here.

      PS. Is VMD assigning correctly the bonds for you lipids molecules? If not, you would need to generate a PSF from your gromacs topology.

       

      Last edit: Ismael Rodriguez Espigares 2018-03-27
  • Physics

    Physics - 2018-03-27

    Thanks for the speedy response.

    Yes, that seems to be the case - apparently some atom names have been altered in the tails of SOPS such that they no longer follow conventional CHARMM atom names.
    I have generated a .psf for it and now it works as normal.

    Well spotted. Thanks so much again.

     

Log in to post a comment.

Want the latest updates on software, tech news, and AI?
Get latest updates about software, tech news, and AI from SourceForge directly in your inbox once a month.