Pdb number of residue limitation 10000

Hi @amjjbonvin:
Is there any way to go around the limitation of residues in the PDB files (10,000)? or the 100,000 total number of atoms? for example, using PDBx/mmCIF format?

Not really I am afraid… Or use different chains for different parts of your molecule, but then you need to keep those as one unit (e.g. with fix origin).

But do you need to represent the entire molecule? If you have some idea about the possible interaction surface, you could remove the irrelevant parts. Those only make the computations heavier.

@amjjbonvin, cool suggestions. I am trying to use one and make a pdb containing as much as possible proteins to overcome this problem (for now). however, when I am trying to make a combination of proteins containing 9996 residues (pretty close to the 10k limit) I am getting this error at the end of the /begin/generate_1.out :

%RTFRDR-ERR: duplicate (P-)RESIdue name CGAN
 %RTFRDR-ERR: duplicate (P-)RESIdue name CGAN
 RESIDUE>        add angle $pat1 $pat2 $pat3
 %RTFRDR error encountered: exceeded MXRTT (RTF) parameter --> recompile program
   (CNS is in mode: SET ABORT=NORMal END)
 *****************************************************
 ABORT mode will terminate program execution.
 *****************************************************
 Program will stop immediately.
          ============================================================
           Maximum dynamic memory allocation:    10755784 bytes
           Maximum dynamic memory overhead:          1040 bytes
           Program started at: 20:13:32 on 14-Mar-2022
           Program stopped at: 20:16:40 on 14-Mar-2022
           CPU time used:     186.6619 seconds
          ============================================================

I wasn’t able to find any way around it. Do you have any suggestions?
BTW, the file size is huge. It is around ~1.8G while for a normal protein, it is just 35M.

You are hitting an array limit in CNS…

Did you compile it with the routines provided with haddock? If not try that first.

If yes, search in the CNS source directory for MXRTT and increase the value. It must be defined in a .inc file.

@amjjbonvin, I will look into MXRTT and try to fix it.
I did compile it with the routines provided with haddock. But since I had some compiling issues regarding the compiler, I just sent an email to you (haddock.support@gmail.com), asking for the executable.
Best,
Mohammad