Reply To: Docking an iron atom to a peptide

[Anonymous]

Hi smlewis,

Thanks for getting back to me! My response is broken up into two pieces, which may seem disjointed. But here goes anyway!

Firstly, I am trying to modify the metalloprotein_abrelax demo to use my peptide sequence, rather than the protein sequence included in the original demo. Thus far, I’m not too worried if the predicted structure is bound to Zn or Fe (I’ll worry about this later!). Here is my sequence:

DTHFPICIFC[CYZ]C[CYZ]GCCHRSKC[CYZ]GMC[CYZ]CKTZ[ZN]

I replaced the fragment files included with the demo with fragment files that I generated using the Robetta fragment server. These fragment files were generated using the non-metal bound primary sequence:

DTHFPICIFCCGCCHRSKCGMCCKTZ

I also replaced the .psipred_ss2 file in the demo with a .psipred_ss2 file produced by the Robetta fragment server. I also modified the flags file to point to these files.

I ran the AbinitioRelax application using this flag file. However, the code crashed with the warning and error:

core.pack.task: Packer task: initialize from command line()
Warning: Unable to locate database file Dunbrack02.lib.bin
ERROR: expected to read 18 libraries from Dun02, but read 10
ERROR:: Exit from: src/core/scoring/dunbrack/RotamerLibrary.cc line: 847

Could you tell me what I’m doing wrong? Why does the code expect to read 18 libraries rather than 10? Is this because of my choice of fragment libraries?

Thanks! and all the best,

–Buz