AMBER Archive (2005)

Subject: Re: AMBER: Making topology files of big moleucles in antechamber

From: Melinda Layten (mlayten_at_gmail.com)
Date: Thu Mar 17 2005 - 13:48:39 CST


Since a quick google search defined a dendrimer as a branched polymer,
you can create "residues" based on the polymer subunit instead of
trying to define the whole molecule in one go. You can then build up
the subunits using leap.

On Thu, 17 Mar 2005 11:09:06 -0500, Hwankyu Lee <leehk_at_umich.edu> wrote:
> Dear amber-users,
>
> I'm trying to make topology of big molecules(dendrimer;atom numbers > ~5000) by
> using antechamber. Although antechamber is suitable for small molecules,
> antechamber is the only tool to create topology files for amber package. So,
> I'm trying to make topology of dendrimer by following manuals and searching
> amber forum.
>
> Some answers in the forum suggested to increase MAXATM in define.h and
> divcon.dim and use '-j 5' and so on.
> So, I increased some parameters in define.h like below.
> --------------------------
> # define COLORTEXT "YES"
> # define REDUCEFACTOR 1
> # define PI 3.1415926
> # define MAXCHAR 6000
> # define MAXATOM 6000
> # define MAXBOND 6000
> # define MAXRING 1000
> # define MAXGAS 500 /*maximum gasiteger parameters*/
> /*
> For MAXRING, no dynamic memory applied since the actuall number is determined
> using a recursive function. However, for small and middle-sized molecules,
> it is unlikely that the ring num is larger than 1000
> */
> # define MAXCYCLE 100
> # define OUTPUTSTEP 10
> # define MAXTWIST 10
> # define ECSLONG 2
> # define COSCUT 120
> # define DEGRAD 3.1415926/180
> # define VDWIDIST 10
> # define ESIDIST 14
> # define THETACUT 15
> # define CUBE 2.0
> # define MAXWILDATOM 20
> # define MAXSCHAIN 100
> # define MAXCES 20
> # define MAXBEED 20
> # define MAXATOMTYPE 250
> # define MAXVASTATE 6000
> # define PSCUTOFF 7
> # define MAX_CES_BOND 100
> ---------------------------
> Also, in divcon.dim file, I modified that into "MAXATM=6000".
> The parametes in divcon.dim are shown in below.
> --------------
> MAXIMUM NUMBER OF ATOMS:
> PARAMETER (MAXATM =6000)
> C
> C FRACTION OF ATOMS THAT ARE EXPECTED TO BE HEAVY:
> PARAMETER (FHEAVY = 0.50)
> C
> C MAXIMUM NUMBER OF ATOM PAIRS (INFINITE CUTOFF):
> PARAMETER (MXPAIR =50000)
> C
> C MAXIMUM NUMBER OF ATOM PAIRS THAT ARE STORED FOR OFF-DIAGONAL
> C INTERACTIONS IN FOCK, 1-ELECTRON, AND DENSITY MATRICES:
> PARAMETER (MBPAIR =400000)
> C
> C MAXIMUM NUMBER OF RESIDUE-BASED ATOM PAIRS (USED FOR AUTOMATIC
> C SUBSETTING, 5 ANGSTROM CUTOFF) (CODE=M):
> C [ SET TO ONE IF CLUSTER-BASED SUBSETTING IS NOT USED ]
> PARAMETER (MXPRES = 10000)
> C
> C MAXIMUM NUMBER OF SUBSYSTEMS:
> PARAMETER (MAXSUB =600)
> C
> C STORAGE FOR SUBSYSTEM ATOM LISTS:
> PARAMETER (MSLIST =350000)
> C
> C STORAGE FOR ALL SUBSYSTEM EIGENVALUES (CODE=L):
> PARAMETER (MSVAL =350000)
> C
> C SCRATCH STORAGE FOR MAXIMUM NUMBER OF ORBITALS IN A SUBSYSTEM (CODE=L)
> C [SEE ALSO MSRORB PARAMETER]:
> PARAMETER (MSORB =2000)
> ----------------------------------
>
> Then, when I used "antechamber -i XXX.pdb -fi pdb -o XXX.prepin -fo prepi -c bcc
> -s 2 -j 5 -nc 128", it took so long time, and then show this error like below.
> ---------------------
> db -o g5ace80.prepin -fo prepi -c bcc -s 2 -j 5 -nc 128
> Running: /home/leehk/local/amber8/exe/bondtype -i ANTECHAMBER_BOND_TYPE.AC0 -o A
> NTECHAMBER_BOND_TYPE.AC -f ac -j part
>
> Running: /home/leehk/local/amber8/exe/atomtype -i ANTECHAMBER_AC.AC0 -o
> ANTECHAMBER_AC.AC -p gaff
>
> Total number of electrons: 18094; net charge: 128
>
> Running: /home/leehk/local/amber8/exe/divcon
> Unable to find mopac charges in divcon.out
> ------------------------------
>
> When I used antechamber with small molecules, it nicely worked. So, I think that
> there is no problem with antechamber in my computer. So, it looks like this
> problem comes from size of molecules.
> How can I use antechamber for big molecules?? If it is impossible to do that,
> how can I generate topology of big molecules? If I cannot generate topology of
> big molecules, is it impossible to simulate big molecules in amber package?
>
> Thanks for your help in advance.
>
> best,
> Hwankyu.
> -----------------------------------------------------------------------
> The AMBER Mail Reflector
> To post, send mail to amber_at_scripps.edu
> To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu
>
-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu