I am trying to use Rosetta for antibody homology modeling and docking.
The method I used was from article："Modeling and docking of antibody structures with Rosetta" (https://www.nature.com/articles/nprot.2016.180)
The Rosetta version I used is: 3.13 Release date: Tuesday, June 1, 2021
I performed antibody homology modeling and docking preparation according to the method in the article. When I run snugDock, the software keeps saying something wrong:
[ ERROR ] UtilityExitException ERROR: Sequence for partner1: EVQLVESGGGLVQPGRSLRLSCAASGFTFDDYAMHWVRQAPGKGLEWVSAITWNSGHIDYADSVEGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCAKVSYLSTASSLDYWGQGTLVTVSDIQMTQSPSSLSASVGDRVTITCRASQGIRNYLAWYQQKPGKAPKLLIYAASTLQSGVPSRFSGSGSGTDFTLTISSLQPEDVATYYCQRYNRAPYTFGQGTKVEIKRT does not match first member of ensemble1: DIQMTQSPSSLSASVGDRVTITCRASQGIRNYLAWYQQKPGKAPKLLIYAASTLQSGVPSRFSGSGSGTDFTLTISSLQPEDVATYYCQRYNRAPYTFGQGTKVEIKRTEVQLVESGGGLVQPGRSLRLSCAASGFTFDDYAMHWVRQAPGKGLEWVSAITWNSGHIDYADSVEGRFTISRDNAKNSLYLQMNSLRAEDTAVYYCAKVSYLSTASSLDYWGQGTLVTVS AN INTERNAL ERROR HAS OCCURED. PLEASE SEE THE CONTENTS OF ROSETTA_CRASH.log FOR DETAILS.
The code I am executing:
docking_prepack_protocol.mpi.linuxgccrelease \ -in:file:s model-0.relaxed_0168.antibody_antigen_start_new.pdb \ -ex1 \ -ex2 \ -partners LH_A \ -ensemble1 antibody_ensemble.1.list \ -ensemble2 antigen_ensemble.list \ -docking:dock_rtmin
Based on the log message, I thought there might be a problem with the chain order in my partner1 file or ensemble1. But after checking, I found that the antibody sequence in the ppk file of ensemble1 is fine, the order is: L chain->H chain. in partner1.ppk file, the order of the chain is fine, also L chain->H chain, but the identifier is changed to A and B, which may be handled by the software docking_prepack_protocol.
When I searched for related questions in the forum, I found that the same question was mentioned before by
So I tried to swap the order of L chain and H chain in the partner1.ppk file and changed the chain identifiers to H and L. The rest of the file remained unchanged.
I reran the above command and it works!
However I'm even more confused!
1, why does snugdock think my sequence is not match?
2, Can I get normal results with the way I'm running it now? Why I set "-partners LH_A" in the parameter and the sequence order of partner1.ppk is H chain->L chain, but snugdock works well?
3, why the sequence identifier in partner1.ppk be changed to A and B after prepack run?