Dear Amber Users,
I am trying to incorporate a virtual spin label into a protein following the approach done by Islam et. al. (S. M. Islam, R. A. Stein, H. S. Mchaourab, B. Roux, J. Phys. Chem. B 2013, 117, 4740–4754.) using leap. I will then solvate the system and use the .prm and .mdcrd files as input for MD in OpenMM. In order to allow for multiple spin labels in a protein, and to better keep track of them, I’ve created new definitions for each residue that contain the virtual spin label (OND) bound to the Calpha. I have modified the leaprc and frcmod files for the f14SB force field, and created a lib file for the new residue definitions (for testing purposes, the lib file is only a modified alanine named ONA). The modifications to the leaprc file are:
-------------------------------------------
addAtomTypes {
...
{“OND”, “ON”, “sp3”} //new atom type
…
frcmod14SB = loadamberparams frcmod.ff14SB_OND //changed the frcmod file
…
loadOff ala_OND.redq.lib //new lib file
…
-------------------------------------------
The changes to the frcmod file resemble those in the Islam et. al. paper (see SI). We treat it very similar to an oxygen:
-------------------------------------------
MASS
ON 16.00 0.000
BOND
CX-ON 1.2 8.0
ANGL
C -CX-ON 2.0 46.0
CT-CX-ON 2.0 46.0
DIHE
N-CX-C-ON 1.9 1 240.0
N-CX-CT-ON 1.9 1 240.0
NONB
ON 2.0000 0.0500
-------------------------------------------
The new lib file is a modification of the amino12.redq.lib file (for testing it only contains a modified alanine residue). The relevant lines are:
-------------------------------------------
!entry.ONA.unit.atoms
...
“OND” “ON” 0 1 131072 11 0 0.0
!entry.ONA.unit.atomsperinfo
...
“OND” “ON” 0 -1 0.0
!entry.ONA.unit.connectivity
...
3 11 1 //CA and OND
!entry.ONA.unit.positions
...
2.552251 8.382596 6.463704
-------------------------------------------
The positions were determined using the following equations:
direction = (ca_pos – cb_pos) / np.linalg.norm(ca_pos – cb_pos)
ond_pos = (direction + ca_pos – n_pos)
When I try to create an alanine chain with a modified residue from sequence, I get a fatal error related to chirality:
-------------------------------------------
Welcome to LEaP!
(no leaprc in search path)
> source leaprc.ff14SB_OND.redq
----- Source: ./leaprc.ff14SB_OND.redq
----- Source of ./leaprc.ff14SB_OND.redq done
Log file: ./leap.log
atom type OND - unknown element ON
Loading parameters: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/parm/parm10.dat
Reading title:
PARM99 + frcmod.ff99SB + frcmod.parmbsc0 + OL3 for RNA
Loading parameters: ./frcmod.ff14SB_OND2
Reading force field modification type file (frcmod)
Reading title:
ff14SB protein backbone and sidechain parameters
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/amino12.redq.lib
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/aminoct12.redq.lib
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/aminont12.redq.lib
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/nucleic12.redq.lib
Loading library: ./ala_OND.redq.lib
.database.c 515|WARNING: Nonunique entry in database: !index found
.database.c 515|WARNING: Nonunique entry in database: entry.ON.unit.connect found
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/atomic_ions.lib
Loading library: /home/kari.gaalswyk/anaconda2/bin/../dat/leap/lib/solvents.lib
> foo = sequence {ALA ALA ALA ONA ALA ALA}
ERROR: Comparing atoms
.R<ONA 4>.A<C 9>,
.R<ONA 4>.A<CB 5>,
.R<ONA 4>.A<HA 4>, and
.R<ONA 4>.A<N 1>
to atoms
.R<ONA 4>.A<C 9>,
.R<ONA 4>.A<CB 5>,
.R<ONA 4>.A<OND 11>, and
.R<ONA 4>.A<N 1>
This error may be due to faulty Connection atoms.
!FATAL ERROR----------------------------------------
!FATAL: In file [chirality.c], line 142
!FATAL: Message: Atom named OND from ONA did not match !
!
!ABORTING.
-------------------------------------------
I get the same error if I try to load a pdb file with the modified residue already incorporated (both with and without the OND atom present). From my understanding, this is because I have defined a pentavalent carbon (by bonding it to the Calpha in the lib file), and the chirality functions are attempting to match the 4 neighbors of the CA, but it now has 5. Is there any way to allow for pentavalency in tleap, or am I going about this the wrong way? I’m not too familiar with amber’s inner workings.
Thank you,
Kari Gaalswyk
_______________________________________________
AMBER mailing list
AMBER.ambermd.org
http://lists.ambermd.org/mailman/listinfo/amber
Received on Thu Jan 10 2019 - 08:00:05 PST
