ERROR READING PDB FILE

[Anonymous]

I am new to rosetta3.0. As i am trying to perform denovo modelling of RNA molecule my script is giving an error again and again like this:

ERROR: unrecognized aa r
ERROR:: Exit from: src/core/io/pdb/file_data.cc line: 476

I have already changed the PDB file format so that res name reads rA, rG, rU, and rC. But still i am getting this message.Can someone suggest whats going wrong?

[Anonymous]

As you suggested, Rosetta isn’t quite happy with the way you’ve formatted the RNA PDB.

For some reason, RNA appears to not be in 3.2, but it was in 3.1 (and I guess 3.0).

Do you have an rna integration test in the folder rosetta_source/test/integration/tests/(rna something?) You can compare that RNA PDB to your own RNA pdb to see what format Rosetta expects. You can also examine the RNA residue types; I don’t know where they are in your copy of the database but it will be something like rosetta_database/chemical/residue_type_sets/(rna?)/types/*.params.

[Anonymous]

Thanks for the input, but i have checked these *.params file mentioned above. And as far as PDBs go then i am providing a sample of both PDBs in-built(rosetta3.0) and my modified: –

In-built PDBs(in rosetta_source)
ATOM 374 C5 rG A 12 -15.576 14.697 9.442 1.00 0.00
ATOM 375 C6 rG A 12 -15.232 15.983 9.940 1.00 0.00
ATOM 376 O6 rG A 12 -14.494 16.852 9.424 1.00 0.00
ATOM 377 N7 rG A 12 -15.194 14.038 8.278 1.00 0.00
ATOM 378 C8 rG A 12 -15.796 12.884 8.340 1.00 0.00
ATOM 379 N9 rG A 12 -16.560 12.732 9.493 1.00 0.00
ATOM 380 1H5* rG A 12 -17.361 8.686 7.054 1.00 0.00
ATOM 381 2H5* rG A 12 -18.945 9.217 6.477 1.00 0.00
ATOM 382 H4* rG A 12 -18.863 9.316 8.825 1.00 0.00
ATOM 383 H3* rG A 12 -19.038 12.072 7.562 1.00 0.00
ATOM 384 1H2* rG A 12 -19.157 12.908 10.021 1.00 0.00
ATOM 385 2HO* rG A 12 -19.211 11.371 11.694 1.00 0.00
ATOM 386 H1* rG A 12 -17.132 11.318 10.908 1.00 0.00
ATOM 387 H1 rG A 12 -15.631 17.118 11.597 1.00 0.00
ATOM 388 1H2 rG A 12 -17.864 15.276 13.509 1.00 0.00
ATOM 389 2H2 rG A 12 -16.992 16.762 13.339 1.00 0.00
ATOM 390 H8 rG A 12 -15.705 12.110 7.562 1.00 0.00
TER

1RNK.pdb(modified):
ATOM 1082 O2 rU A 34 -25.519 -3.185 11.349 1.00 0.00 O
ATOM 1083 N3 rU A 34 -26.656 -5.031 12.022 1.00 0.00 N
ATOM 1084 C4 rU A 34 -27.107 -5.927 12.966 1.00 0.00 C
ATOM 1085 O4 rU A 34 -27.806 -6.874 12.623 1.00 0.00 O
ATOM 1086 C5 rU A 34 -26.688 -5.642 14.319 1.00 0.00 C
ATOM 1087 C6 rU A 34 -25.911 -4.561 14.575 1.00 0.00 C
ATOM 1088 H5′ rU A 34 -24.511 -1.035 17.805 1.00 0.00 H
ATOM 1089 H5” rU A 34 -25.539 -2.234 16.991 1.00 0.00 H
ATOM 1090 H4′ rU A 34 -22.762 -1.409 16.072 1.00 0.00 H
ATOM 1091 H3′ rU A 34 -24.945 0.209 15.679 1.00 0.00 H
ATOM 1092 HO3′ rU A 34 -23.937 0.448 13.434 1.00 0.00 H
ATOM 1093 H2′ rU A 34 -26.340 -1.551 14.787 1.00 0.00 H
ATOM 1094 HO2′ rU A 34 -26.076 -1.216 12.356 1.00 0.00 H
ATOM 1095 H1′ rU A 34 -23.980 -2.361 13.022 1.00 0.00 H
ATOM 1096 H3 rU A 34 -26.928 -5.198 11.066 1.00 0.00 H
ATOM 1097 H5 rU A 34 -26.997 -6.296 15.134 1.00 0.00 H
ATOM 1098 H6 rU A 34 -25.608 -4.357 15.602 1.00 0.00 H
TER 1099 rU A 34
MASTER 46 0 0 0 0 0 0 6 1098 1 0 3
END

Now i dont know how much is the difference between the two files.

[Anonymous]

A) The spacing is wrong in both PDBs. I think this is just an artifact of cutting and pasting into this forum software. PDBs are fixed-column formats, so (for example) all the H’s on the right should line up. (Um, they aren’t all hydrogen, why all H?) It’s possible this is the problem, if it thinks there is a residue named “r”, when clearly they’re rA/rU/etc in your example.

The atom names are totally different. 1H5* becomes H5′ (?) If this is restricted to just the hydrogens, it ought not to cause a problem. If heavyatoms are misnamed, it will cause Rosetta to reject those residues.

C) Which executeable are you using? If it’s not one of the rna_ apps, we might need to hack the database a bit (my coworker did, anyway)…

[Anonymous]

I don’t think that there are any heavy atoms in the PDB file. The executables which i am using is :- rna_denovo.linuxgccrelease.

And in rosetta PDB files H’ are marked as H* and the numbering of hydrogens do differ like H5′ is 1H5* and H5” is 2H5*.If this is the problem then the PDB files need heavy modifications??(i think)..Is there any way to modify *params files or any other file in source code itself to make things a bit less complicated?

[Anonymous]

yup,the PDB file is 1RNK.pdb which i have modified like this:

[Anonymous]

It looks like you have one too many columns between the atom name column and the residue name column.

Also, your C and O atoms have ‘ instead of *, which may or may not cause a problem.

I attached an example of your spacing versus the correct spacing.

[Anonymous]

“Heavy atoms” in this context means “anything heavier than hydrogen”.

Try attaching your PDB (there should be a file attachment button just above “preview” when you reply) so I can look at it without the forum corrupting it.

[Author]

Posts