Opening a PDB file shows double H atoms

Hi,


I got a PDB file from Cambridge Crystallographic Data Center using the downlaod program ConQuest. Opening this PDB file with Marvin Space produces the picture shown in the screenshot366. There are two hydrogen atoms connected to the benzene rings and and the amid-nitrogens. Opening the same file with Jmol shows the molecule with the normal hydrogens which are given in the file, see screenshot366. I controlled the PDB-file with a text editor, the number of hydrogens is OK.





Converting the file with the program babel from ACTHBZ.pdb to ACTHBZ.xyz, opening this xyz-file with Marvin Space, Marvin Space shows the molecule correctly with only one hydrogen at all atoms (and of course normal methyl groups).





Why are in the first display "double" hydrogens ?





Regards, Hans-Ulrich

Hi Hans-Ulrich,





these weird hydrogen atoms are related to a bug in hybrodization and valence recognition in PDB import. We will fix this high priority bug in the next release (version 5.0.2) of Marvin. You'll be notified when it's available.





Thank you for reporting this problem and apologies for any inconvenience this might cause.





Regards,


Miklos

Hi Miklos,





why to calculate the hybridization when the H-atoms are all present and the number of neighbors of all atoms is defined ?





Now I wonder if it is a bug in MarvinSpace. I tried to look other proteins coming directly from the protein data bank, f.e. 1OEH.pdb, and the H-atoms are all OK in MarvinSpace.


Could it be a failure in the ACTHBZ.pdb file ? I have downloaded it from the CCDC data bank (because it's not a protein) and converted it to a pdb-file using the download program ConQuest, see screenshot.


I will attach the pbd-files 1OH.pdb and ACTHBZ.pdb, so you can have a look at these pdb-files. I did not find a failure in it.





Best regards, Hans-Ulrich

Hi Miklos,


I analyzed further these two pdb files and I found:





For ACTHBZ.pdb the lengths of "aromatic" C-H bonds are different in Jmol and in MarvinSpace. Jmol shows values from 0.92 to 1.06 A. MarvinSpace shows all bondlenghts as 1.09 (for both H-atoms attached to an "aromatic" C).





For 1OEH.pbd both programs show for the "aromatic" C-H the same the C-H bonds lengths (all 1.08 A)





Does MarvinSpace "add" the H-atoms in ACTHBZ.pdb ?





Regards, Hans-Ulrich

Hi Hans-Ulrich,





apologies for some delay in responding to your enquiry. I will get back with specific and detailed answers soon.





Thank you for your patience.





Regards,


Miklos

Hi Hans-Ulrich, this bug has been fixed. The new release of marvin (5.0.2) should be available within days. You'll be notified.


Thank you for the bug report.





Regards,


Miklos