BUSTER Output Interpretation

 

Examine the refine log file

  • Always capture the standard output of refine to a log file.
    • the normal way to do this is to use the > redirect symbol when starting refine
           % refine -p some.pdb -m other.mtz -d results.dir > refine.log
    • if you prefer it is also possible to use tee so you can see the output in the terminal while at the same capturing it too:
           % refine -p some.pdb -m other.mtz -d results.dir | tee refine.log
  • The refine.log file reports important information. It can be examined in an editor or with less or by using grep.
    • if the job fails it will contain an error message (often this points to additional information in separate log-files: make sure to always follow the information given in the ERROR messages)
    • if the refinement runs through, it is recommended to still look for lines containing the string "WARNING". WARNING messages can often be accepted, but for instance if you re-refine pdb entry 1udt with a grade dictionary for viagra a warning is generated:
            % grep WARNING 1udt_refine.log
WARNING: Have |delta/sigma| deviations > 5.0 sigma. Number of outliers for each term: 
WARNING:        12 bond lengths. Worst is    16.7 sigs     1.31 Angs A|1000:S10=O12 (VIA)
WARNING:        13 bond angles.  Worst is    16.5 sigs   105.07 degs A|1000:O11=S10=O12 (VIA)
WARNING:         2 idealD contacts. Worst     6.0 sigs     2.56 Angs A|677:SG=SG (CYS) symm: 1555=3556
WARNING:         1 Bcorrel.      Worst is     5.6 sigs    22.34 Angs**2 A|590:SD=CE (MET)
      • in this case the VIA residue has some unusual bond angles and bond lengths which can be fixed by refinement. More interestingly is the contact between atom SG of CYS A 677 and a symmetry copy of itself. Looking at the this atom in the final structure and map reveals that a disulfide bond is likely to have been formed.
    • of particular importance are WARNING messages at the end of the refinement. This is because refinement fails to remove them.

The output also reports "table 1" type information:

rms bond deviation should be in range 0.008 to 0.012 Å
rms angle deviation should be around 1.0 to 1.4 degrees
Rwork Rfree depends

return to BUSTER Output Interpretation


Page by Oliver Smart 07 Feb 2011. Any questions regarding our software or this wiki should be directed to buster-develop@globalphasing.com