AMBER Archive (2004)

Subject: Re: AMBER: problems in using glycam

From: Wen-Chi Tseng (tsengwc_at_ch.ntust.edu.tw)
Date: Thu Jul 01 2004 - 07:44:27 CDT

> > 3. I decided to construct a simple sugar, maltose, using
> > glycam04. Here is what I did and what I got. It seems that I need
> > to define the torsions and charge of the molecule. Can anyone
> > advise how to do this and where to get the correct torsion
> > information?
> >
> > mal = sequence { 1GA 4GA }
> > saveamberparm mal mal.top mal.crd
> >
> > Checking Unit.
> > ERROR: The unperturbed charge of the unit: 0.199000 is not
> > integral. WARNING: The unperturbed charge of the unit: 0.199000
> > is not zero.
>
> This is because our carbohydrates are all reduce - they have a
> methoxy functional group at the C1 position instead of a hydroxy
> group. Adding the methoxy group. For the alpha anomer of maltose
> it would be mal = sequence { OME 4GA 1GA).
>
> These will go away with the correct build sequence { OME 4GA 1GA }.
>
> Hope this helps and clarify things. Here is the glycam04 build
> sequence for leap:
>

Thank you for your advise. I have new questions for using glycam.

Do I need to add the methoxy group to all the non-bonded C1 positions
of any carbohydrate constructed by glycam? Or just adding to the
terminal one?

If I want to build trehalose, could you advise where to put the
methoxy group?

If I want to use the carbohydrate built by glycam in a docking
program, such as Autodock, will the extra methoxy group affect the
docking process?

I used vmd to examine the maltose built by glycam04 after I saved the
molecule in pdb file. However, I could not see the ring structure of
the molecule. Did I do something wrong?

If I can obtain a pdb file of the carbohydrate in crystal structure
instead of using glycam to build one, how should I modify the pdf
file to do MD in Amber?

I have another questions about the new version of the building
sequence. (see below)

> logFile leap.log
> #
> #
> # Making Branched Carbohydrates
> #
> # In general, the head atom is set to C1 in each sugar, where the
> tail atom
> # varies depending on the sugar linkage. For all sugars, the tail
> atom is
> # specified as the glycosidic oxygen with the largest atomid, i.e.
> for the
> # 2,3,4, linked sugars the tail atoms would be O4. Therefore, it
> it possible
> # to build the longest chain according to this convention and then
> connect
> # the branched portions. Simply changing the tail atom enables a
> different
> # connection point for the unit, which can then be utilized to
> sequence to
> # any branched residue/s.
> #
> #
> # Example: Man9
> #
> # Mana1 - 2Mana1 \
> # 3
> # Mana1
> # 6 \
> # Mana1 - 2Mana1 / 6
> # ManB-OMe
> # 3
> # Mana1 - 2Mana1 - 2Mana1 /
> #
> #
> # First, we must load the necessary parameters and prep files
> #
> source leaprc.Glycam_04
> #
> # Find the longest chain and use the sequence command to build it
> # (See the manual for naming conventions)
> #
> part1 = sequence { OME VMB 2MA 2MA 1MA }
> #
> # Now set the tail atom of part1 to O3 of VMB
> #
> set part1 tail part1.2.9

I am confused. Does this command reassign the number sequence of
atoms? When I used "desc part1.2" to check the atom numbers, the atom
of part1.2.9 is O6, not O3, and the O3 atom of VMB is part1.2.20.

> #
> # Join the first branch to the long chain
> #
> part2 = sequence { part1 VMA 2MA 1MA }
> #
> # Set the tail atom of part2 to O3 of VMA
> #
> set part2 tail part2.6.19

I am confused again. Same reason: the molecule of part2.6.19 is O6,
not O3.

> #
> # Add the last branch
> #
> man9 = sequence { part2 2MA 1MA }
> #
> # The basic structure has been built, but it clearly does not have
> the optimal
> # glycosidic torsion angles.
> #
> # Set the psi(1-6) torsion to 180 (C1-O6-C6-C5)
> #
> impose man9 { 2 3 4 } { { C5 C6 O6 C1 180.0 } }
> #
> # Set the omega angle of 6-linked sugars to 60.0 (O5-C5-C6-O6)
> #
> impose man9 { 2 3 } {{ O6 C6 C5 O5 60.0 } }
> #

I got segmentation fault here, and cannot go on any further.

> # Set the phi angle of all a-linked (not 1-6 linked) sugars to
> -60.0 #
> impose man9 { 4 5 6 7 8 9 10 } { { H1 C1 O2 C2 -60.0 } }
> impose man9 { 6 2 } { { H1 C1 O3 C3 -60.0 } }
> impose man9 { 9 3 } { { H1 C1 O3 C3 -60.0 } }
> # Set the psi angle of all sugars to 0.0
> #
> impose man9 { 4 5 6 7 8 9 10 } { { C1 O2 C2 H2 0.0 } }
> impose man9 { 6 2 } { { C1 O3 C3 H3 0.0 } }
> impose man9 { 9 3 } { { C1 O3 C3 H3 0.0 } }
> #
> # Now we have a reasonable starting structure for AMBER min/md, but
> we need
> # to save a topology and coordinate file as well as a pdb for later
> use. #
> saveamberparm man9 man9.top man9.crd
> savepdb man9 man9.pdb
> #
> # Exit the program
> #
> #quit
> #
> # For more info visit the GLYCAM website www.glycam.ccrc.uga.edu
>
> Cheers,
> Karl
>
> ---------------------------------
> Karl Nicholas Kirschner, Ph.D.
> Assistant Professor of Chemistry
> Hamilton College, NY 1323
> ---------------------------------
>

Thank you very much.

Wenchi

.

-----------------------------------------------------------------------
The AMBER Mail Reflector
To post, send mail to amber_at_scripps.edu
To unsubscribe, send "unsubscribe amber" to majordomo_at_scripps.edu