Known Issues in 20190607 BUSTER release:


Please also check the BUSTER home page and the news page for updates and snapshot releases. Issues mostly related to external programs can still be active: please check here.

mmCIF file deposition problem (fixed in 20190923 release)

When an mmCIF file produced by GEMMI is submitted to OneDep or to the validation server, the following errors are shown:

Coordinate issues

ERROR: Missing '_entity_poly.pdbx_strand_id' item.

ERROR: For '_atom_site.label_asym_id' and '_atom_site.auth_asym_id' fields, only alphanumeric values (A-Z, 0-9, a-z) are allowed.

The underlying reason are different interpretations of the PDBx/mmCIF dictionary: while Gemmi follows the above dictionary, the (different) deposition/validation services provided by wwPDB have stricter requirements.

One possible (temporary) fix for the moment could be:

  • replace the occurences of [A-Z]:[0-9a-z] with a single identifier, e.g. via (for chains A-E and W):
      sed -e "s/ A:[0-9a-z] / A /g" \
          -e "s/ B:[0-9a-z] / B /g" \
          -e "s/ C:[0-9a-z] / C /g" \
          -e "s/ D:[0-9a-z] / D /g" \
          -e "s/ E:[0-9a-z] / E /g" \
          -e "s/ W:[0-9a-z] / W /g" \
          BUSTER_model.cif > BUSTER_model_new.cif
  • remove the whole "_entity_poly" loop from the resulting BUSTER_model_new.cif file (via an editor)

You could try running the script to do these steps for you:

      chmod 0755 /where/ever/
  • run this script inside the BUSTER output directory (ie. where the BUSTER_model.cif file is):
  • use the resulting *_new.cif files for deposition

Updated Gemmi binary (20190830)

Updated Gemmi binaries are attached to this page:

Furthermore additional parts are now included in the model.cif file especially some "refine_ls_restr" components. After downloading you can use the updated Gemmi binary either in refine:

    env autoBUSTER_Exe_gemmi=/path/to/gemmi-20190830 refine ...

or in aB_deposition_prep:

    env autoBUSTER_Exe_gemmi=/path/to/gemmi-20190830 aB_deposition_prep BUSTER_model.pdb BUSTER_refln.mtz

Deposition problem - pdb_extract

When using the online PDB_EXTRACT service for preparation of deposition files or when PDB_EXTRACT is used as part of the deposition process, several issues have been observed.

The program doesn't like comments at the end of lines and will produce (e.g. in the log sf_convert log file) messages like

Warning: File (/data/.../dir_2019_7_16_11_3728/sf_for_refine.hkl) has item (_refln.index_h                    # H              from dataset HKL_base) not in the pdbx dictionary.

This doesn't seem to prevent the validation/deposition software from accepting the BUSTER reflection mmCIF file though - but please check if all data blocks and reflection items make it into the deposited structure factor file.