Error in minimization step “Some cmap parameters are missing”

Viewing 3 reply threads
  • Author
    Posts
    • #25920

      BTBTIU16002
      Participant

      Dear GENESIS developers and users,

      I faced this error when performing the minimization step for my protein-membrane system:

      Setup_Enefunc_Cmap> Natural spline is used to derive cmap coefs.

      Setup_Enefunc_Cmap> Some cmap parameters are missing. rank_no = 0

      I am not sure if this is a technical issue of GENESIS or something related to the setting up. Could you please help me explain the case and suggest me any solution to this problem?

      Thank you so much for your time and consideration.

      Yours sincerely,

      Dung

    • #25922

      ckobayashi
      Moderator

      Hello,

       

      The location of the error suggests that the atoms involved in the cmap are not assigned correctly. The error is provided, but this is the first time I have seen a case where this error is actually made. Could you please give me some more information to understand the situation? It would be helpful if you could upload or paste the standard out file if possible.

    • #25954

      BTBTIU16002
      Participant

      Dear Dr. Kobayashi,

      I’m so sorry for my poor description. I would like to paste the content of the out file below to provide you with more information:

      This is the stdout.1.2 file:

      WARNING: too small distances: 9998 10004 0.47392
      WARNING: too small distances: 9998 9999 0.34291
      WARNING: too small distances: 9998 10000 0.48752
      WARNING: too small distances: 10003 10004 0.33257
      WARNING: too small distances: 10003 10001 0.25833
      WARNING: too small distances: 10003 10002 0.34574
      WARNING: too small distances: 10003 10005 0.39950
      WARNING: too small distances: 10003 10009 0.34298
      WARNING: too small distances: 10004 9999 0.41795
      WARNING: too small distances: 10004 10001 0.40219
      WARNING: too small distances: 10004 10005 0.35902
      WARNING: too small distances: 9999 10000 0.21218
      WARNING: too small distances: 9999 10001 0.33289
      WARNING: too small distances: 9999 10002 0.49691
      WARNING: too small distances: 10000 10001 0.42150
      WARNING: too small distances: 10001 10002 0.20012
      WARNING: too small distances: 10002 10009 0.49974
      WARNING: too small distances: 10005 10006 0.20657
      WARNING: too small distances: 10005 10007 0.32900
      WARNING: too small distances: 10005 10008 0.49944
      WARNING: too small distances: 10006 10007 0.37996
      WARNING: too small distances: 10006 10008 0.46151
      WARNING: too small distances: 10007 10008 0.22244
      WARNING: too small distances: 10007 10009 0.43086
      WARNING: too small distances: 10007 10011 0.27385
      WARNING: too small distances: 10007 10012 0.45032
      WARNING: too small distances: 10008 10011 0.39179
      WARNING: too small distances: 10008 10012 0.46815
      WARNING: too small distances: 10009 10010 0.22005
      WARNING: too small distances: 10009 10011 0.28618
      WARNING: too small distances: 10009 10012 0.46845
      WARNING: too small distances: 10010 10011 0.39819
      WARNING: too small distances: 10010 10012 0.48864
      WARNING: too small distances: 10011 10012 0.22693
      Setup_Enefunc_Cmap> Some cmap parameters are missing. rank_no = 2

      Once again, thank you so much for your help.

    • #25955

      ckobayashi
      Moderator

      Hello,

      Are atoms with an atom index around 10,000 lipid molecules or proteins? Could you check some of the two atoms that are shown as short?
      In a normal CHARMM force field (and most other force fields), if the atom indexes are close, they are likely to be in the same molecule.
      If the distance between atoms in the same molecule is unusually short (<0.5 Å or so), then either you have done something wrong in generation of simulation system, or you define the wrong box sizes (box_size_[x,y,z]) in [BOUNDARY].

       

      Sincerely yours,

       

      Kobayashi

Viewing 3 reply threads

You must be logged in to reply to this topic.