After doing ab initio structure prediction on the individual domains of my target protein I was trying to assess the quaternary structure of the dimer using the SymDock protocol. The results I get are, firstly, never for a full-length protein and, secondly, the amino acid numbering is confused; trying to extract pdbs from the silent out file results in an error message (core.io.silent: ERROR: add_chain_ending() invalid chain ending 226
ERROR:: Exit from: src/core/io/silent/ProteinSilentStruct.tmpl.hh line: 671).
The construct I am dealing with looks like this: --[domain A]-----[domain B]
I would like to know whether I am using a correct protocol or how one would use Rosetta to model a dimeric, but multi-domain protein?
This is my run file:
Residue Pair Jump File:
jump_def: 61 262 261 261
aa: ASN FMN
cst_atoms: CA CB CG C4 C5 O7
jump_atoms: N CA C C4 C5 O7
Symmetry Definition File:
E = 2*VRT0001 + 1*(VRT0001:VRT0002)
start -1,0,0 0,1,0 0,0,0
rot Rz 2
connect_virtual JUMP1 VRT0001 VRT0002
set_dof BASEJUMP x(50) angle_x(0:360) angle_y(0:360) angle_z(0:360)
Thank you very much in advance for any advise.
PS: I just tried 3.3 but to that does not change anything.