difficulty using mmcif models in Rosetta

[alebus1967]

Hello, I am having trouble to use mmcif (and not pdb) format models as input to do relaxation with a cryoEM map.
I need to stick to mmcif.

The error is :
"Cannot determine file type. Current supported types are: PDB, CIF, SRLZ, MMTF"

Rosetta (rosetta_scripts.static.linuxgccrelease ) version: rosetta.binary.linux.release-371

Yet, my file is a valid mmcif (used in coot/chimerax, as well as for phenix and servalcat refinement; and also checked using gemmi validate) with suffix .cif
Rosetta treats it as a pdb, and then of course crashes.
The ROSETTA_CRASH.log doesn't give any further detail.

Is there an additional, explicit input option that I need to include in my script? (other than just -in:file:s mycif.cif)
Thank you in advance!

[roccomoretti]

mmcif files should be supported. (When it says supported types include CIF, that means mmCIF.)

The file determination heuristic is based on lines in the file starting with an underscore character. After that, it's checked to see if the mmCIF file parser we use has any diagnostics when parsing. If there's possible issues with the file, we assume it wasn't an mmCIF file. -- This is my guess. There's something the mmCIF parser doesn't like about the file, so it doesn't think it's a valid mmCIF file.

Unfortunately, there doesn't seem to be a way to force loading of a file as mmcif if the autodetection logic doesn't like it. -- If some of the other programs are printing out diagnostic message, see if you can "fix" the CIF file by correcting the warnings.

[alebus1967]

thank you for your reply!
Yet, it is not at all obvious what Rosetta's parser is not liking.

As I said before, I double checked proper mmcif formatting with different tools (gemmi, etc)

Here's a snippet of my mmcif file(s), and below also the Rosetta crash log:

##CIF_1.1

mmCIF

data_all_corr.cif

_audit_conform.dict_name mmcif_pdbx.dic
_audit_conform.dict_version 4.007
_audit_conform.dict_location http://mmcif.pdb.org/dictionaries/ascii/mmcif_pdbx.dic

_audit_syntax.case_sensitive_flag Y
_audit_syntax.fixed_width "atom_site atom_site_anisotrop"

_citation.id chimerax
citation.title
;UCSF ChimeraX: Structure visualization for researchers, educators, and developers
;
<...>
loop
_atom_site.group_PDB
_atom_site.id
_atom_site.type_symbol
_atom_site.label_atom_id
_atom_site.label_alt_id
_atom_site.label_comp_id
_atom_site.label_asym_id
_atom_site.label_entity_id
_atom_site.label_seq_id
_atom_site.Cartn_x
_atom_site.Cartn_y
_atom_site.Cartn_z
_atom_site.auth_asym_id
_atom_site.auth_seq_id
_atom_site.pdbx_PDB_ins_code
_atom_site.occupancy
_atom_site.B_iso_or_equiv
_atom_site.pdbx_PDB_model_num
ATOM 1 N N . MET D4 2 1 -2.307 27.606 -13.741 D4 1 ? 1.00 167.07 1
ATOM 2 C CA . MET D4 2 1 -1.821 26.752 -12.671 D4 1 ? 1.00 167.07 1
ATOM 3 C C . MET D4 2 1 -0.315 26.531 -12.858 D4 1 ? 1.00 167.07 1
ATOM 4 O O . MET D4 2 1 0.155 25.985 -13.857 D4 1 ? 1.00 167.07 1
ATOM 5 C CB . MET D4 2 1 -2.700 25.490 -12.666 D4 1 ? 1.00 167.07 1
ATOM 6 C CG . MET D4 2 1 -2.522 24.456 -11.587 D4 1 ? 1.00 167.07 1
ATOM 7 S SD . MET D4 2 1 -0.876 23.812 -11.459 D4 1 ? 1.00 167.07 1
ATOM 8 C CE . MET D4 2 1 -0.924 23.122 -13.106 D4 1 ? 1.00 167.07 1
ATOM 9 N N . ILE D4 2 2 0.410 26.977 -11.835 D4 2 ? 1.00 162.16 1
ATOM 10 C CA . ILE D4 2 2 1.869 26.944 -11.775 D4 2 ? 1.00 162.16 1
ATOM 11 C C . ILE D4 2 2 2.445 25.738 -11.046 D4 2 ? 1.00 162.16 1
ATOM 12 O O . ILE D4 2 2 2.215 25.551 -9.847 D4 2 ? 1.00 162.16 1
ATOM 13 C CB . ILE D4 2 2 2.412 28.244 -11.168 D4 2 ? 1.00 162.16 1
ATOM 14 C CG1 . ILE D4 2 2 1.836 29.460 -11.903 D4 2 ? 1.00 162.16 1
ATOM 15 C CG2 . ILE D4 2 2 3.908 28.201 -11.095 D4 2 ? 1.00 162.16 1
ATOM 16 C CD1 . ILE D4 2 2 2.216 30.786 -11.290 D4 2 ? 1.00 162.16 1
ATOM 17 N N . ILE D4 2 3 3.192 24.921 -11.779 D4 3 ? 1.00 146.20 1
<...>

##############################################################################################################

Rosetta crash log. Please submit the contents of this file to http://crash.rosettacommons.org/

or use the script Rosetta/main/source/scripts/python/public/report_crashes.py

[START_CRASH_REPORT]
[ROSETTA_VERSION]: 2024.09+release.06b3cf8
[COMMIT_DATE]: 2024-03-01T01:30:53.796737
[APPLICATION]: /programs/x86_64-linux/rosetta/3.14/main/source/bin/rosetta_scripts.static.linuxgccrelease
[MODE]: Release
[EXTRAS]: default
[OS]: GNU/Linux
[COMPILER]: GCC version "4.8.5 20150623 (Red Hat 4.8.5-44)"
[STDLIB]: libstdc++ version 20150623
[START_OPTIONS]
-in:path:database=/programs/x86_64-linux/rosetta/3.14/main/database/ -in:file:s=./Chunks/chunk_001.cif -out:suffix=_relaxed -out:path:all=./Relaxed_chunks/ -parser:protocol=light-constraints.xml -multithreading:total_threads=4 -abinitio:debug -edensity:mapfile=./Chunks/chunk_001.mrc -edensity:mapreso=4.8

[END_OPTIONS]

[0x6bd58c0]
[0x6b2e0f3]
[0x6bdb0a6]
[0x4192d82]
[0x41932bd]
[0x4193476]
[0x40d2091]
[0x40ae10d]
[0x40b0391]
[0x414d518]
[0x41bf5e]
[0x73a6db4]
[0x73a7031]
[0x653ae6]

[END_BACKTRACE]

[START_MESSAGE]
[ ERROR ] UtilityExitException
ERROR: Cannot determine file type. Current supported types are: PDB, CIF, SRLZ, MMTF

[END_MESSAGE]
[END_CRASH_REPORT]
##############################################################################################################

[roccomoretti]

From the information you provided, I can't quite determine why the mmCIF reader isn't liking your file.

However, based on your issue, I came up with PR #356, which is now merged into main. -- Not only should this allow better debugging output about why mmCIF aren't being chosen as such (especially if you add the option -out:levels core.import_pose.util:debug to the command line), it should also fall back on trying to use the filename extension to type the file. -- Of course, if the CIFPARSE-OBJ library that Rosetta uses can't read the mmCIF file for some reason, that doesn't help you. (But at least you'll get a tracer message about why it isn't working properly.)

To use, you'll have to clone this repo locally, and then compile the main branch. -- If you need/want to use a specific older version of Rosetta for your main run, you should be able to compile the development version to test-load your mmCIF files, and get the diagnostic message from the mmCIF parser. Hopefully you can use that to "correct" the mmCIF file for the older version of Rosetta. (The older versions of Rosetta need to have 100% empty diagnostic messages, with not even a single warning or informational message in order to auto-type things as an mmCIF file.)