Disulfide bond coarse-graining using sirah fails

[astomer2]

I saw the article "Protocols for Multi-Scale Molecular Dynamics Simulations in Amber and Gromacs: a Case Study of Intrinsically Disordered Amyloid Beta" from biorxiv. I want to use your method to reproduce the simulation of dynamic disulfide bonds in amber, using pdb2pqr There is no problem in building the pqr file, but there is a problem when using tleap to build the topology file. I used the sirah force field that comes with amber, and also used your latest force field. However, an error occurred when saving the topology file at the end. The following is My error message，I will omit note and warning：

`Error: ** No torsion terms for atom types: GO-GC-QS-QS
for atom GO at position -2.983630, -9.518855, 1.114266,
atom GC at position -1.961885, -11.661592, 1.149774,
atom BSG at position -1.507890, -12.177567, -1.626845,
and atom BSG at position -3.727919, -10.837259, -3.917338.

Error: ** No torsion terms for atom types: QP-QS-QS-GC
for atom BPG at position -2.127612, -11.880519, -2.353026,
atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
and atom GC at position -4.371631, -8.646653, -2.256873.

Error: ** No torsion terms for atom types: QP-QS-QS-QP
for atom BPG at position -2.127612, -11.880519, -2.353026,
atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
and atom BPG at position -4.154724, -10.238984, -4.595233.

Error: ** No torsion terms for atom types: QS-QS-GC-GN
for atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
atom GC at position -4.371631, -8.646653, -2.256873,
and atom GN at position -4.054652, -7.966779, -1.016946.

Error: ** No torsion terms for atom types: QS-QS-GC-GO
for atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
atom GC at position -4.371631, -8.646653, -2.256873,
and atom GO at position -6.148556, -8.597579, -3.837478.

Error: ** No torsion terms for atom types: GC-QS-QS-GC
for atom GC at position -1.961885, -11.661592, 1.149774,
atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
and atom GC at position -4.371631, -8.646653, -2.256873.

Error: ** No torsion terms for atom types: GC-QS-QS-QP
for atom GC at position -1.961885, -11.661592, 1.149774,
atom BSG at position -1.507890, -12.177567, -1.626845,
atom BSG at position -3.727919, -10.837259, -3.917338,
and atom BPG at position -4.154724, -10.238984, -4.595233.

Error: ** No torsion terms for atom types: GN-GC-QS-QS
for atom GN at position -2.665125, -12.129782, 2.341865,
atom GC at position -1.961885, -11.661592, 1.149774,
atom BSG at position -1.507890, -12.177567, -1.626845,
and atom BSG at position -3.727919, -10.837259, -3.917338.
Building improper torsion parameters.
total 32 improper torsions applied
Building H-Bond parameters.
Incorporating Non-Bonded adjustments.

Warning: Parameter file was not saved.`

Judging from the error, it is that some dihedral information is missing in the force field.

[luciannahss]

Greetings, Astomer2.
I apologize for the delay in our reply.

To assist you further, I have a few inquiries regarding your system:

1. Since you are working with S-S bridges, in the pqr file, were the CYS in the S-S bridges designated as CYX, prior to executing the cgconv.pl script?

2. Since leap routinely renumbers the residues, one consequence is that the bond command requires a different position for the sX than the one in the PDB or PQR. Did you check if you are using the correct position for the sX?

3. Does your system contain residues that are missing? Leap attempts to bond segments that are absent; however, if the segments are too far apart, this becomes unattainable. Thus, it is necessary to use a program such as Modeller or Alphafold to model missing segmets.

These are the first thoughts that come to mind, but it could be something else. So, could you kindly provide us with your Leap input and pqr for us to check?