Re: [chimera-dev] [Chimera-users] alignment + missing residues
Dear Eric,
thanks for your suggestions. Unfortunately removing the SEQRES record did not help. I also downloaded and installed the new chimera-version today. The false numbering in the alignment still occurs. Also when I remove the SEQRES record and try with the new chimera version the false numbering appears. Quite annoying... The pdb-file is not in the protein data bank yet.
Any more suggestions? Could I add "dummy atoms" in the structure?
Thanks, Gesa
Eric Pettersen schrieb:
On Jul 14, 2010, at 6:53 AM, Gesa Volkers wrote:
Dear All,
I observed an issue with the alignment function of Chimera. I loaded a crystal structure where in between a loop is not modelled due to low electron density- so these residues are missing in the structure and the numbering also has this gap. When I align this structure with others, the missing residues are not recognized and chimera just overrides the numbering so after this loop all the numbers in the alignment do not fit anymore to the residues they should belong to although the structure and numberings are shown in the right way in the graphics window. How can I overcome this issue? Should I change the pdb-file? In which way?
Thanks for all responses!
Hi Gesa, I had a similar problem reported with 1JFF 12 days ago and haven't yet worked on fixing it. It seems to only happen for a small subset of structures. 1JFF has a missing loop and also exactly one missing residue at the N terminus. Is your structure like that? Is it a standard PDB entry (if so, what's the ID code)? I'll add you to the recipient list for the ticket we have open in our bug database for this (#8585) so you will be notified when it gets fixed. It'll probably be fairly soon. Until then the only workaround that occurs to me is for you to remove the SEQRES records from the PDB file before you open it. Then the Sequence viewer will depict only the residues physically present in the structure -- no gaps (incorrect or otherwise) will be depicted.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
Additionally I downloaded 1JFF and did an alignment with another tubulin-structure. And still the numbering does not recognize this gap between 34 and 61.
Gesa Volkers schrieb:
Dear Eric,
thanks for your suggestions. Unfortunately removing the SEQRES record did not help. I also downloaded and installed the new chimera-version today. The false numbering in the alignment still occurs. Also when I remove the SEQRES record and try with the new chimera version the false numbering appears. Quite annoying... The pdb-file is not in the protein data bank yet.
Any more suggestions? Could I add "dummy atoms" in the structure?
Thanks, Gesa
Eric Pettersen schrieb:
On Jul 14, 2010, at 6:53 AM, Gesa Volkers wrote:
Dear All,
I observed an issue with the alignment function of Chimera. I loaded a crystal structure where in between a loop is not modelled due to low electron density- so these residues are missing in the structure and the numbering also has this gap. When I align this structure with others, the missing residues are not recognized and chimera just overrides the numbering so after this loop all the numbers in the alignment do not fit anymore to the residues they should belong to although the structure and numberings are shown in the right way in the graphics window. How can I overcome this issue? Should I change the pdb-file? In which way?
Thanks for all responses!
Hi Gesa, I had a similar problem reported with 1JFF 12 days ago and haven't yet worked on fixing it. It seems to only happen for a small subset of structures. 1JFF has a missing loop and also exactly one missing residue at the N terminus. Is your structure like that? Is it a standard PDB entry (if so, what's the ID code)? I'll add you to the recipient list for the ticket we have open in our bug database for this (#8585) so you will be notified when it gets fixed. It'll probably be fairly soon. Until then the only workaround that occurs to me is for you to remove the SEQRES records from the PDB file before you open it. Then the Sequence viewer will depict only the residues physically present in the structure -- no gaps (incorrect or otherwise) will be depicted.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
Dear Gesa, I suspect you are using a Mac, since the daily builds failed last night for the Mac platforms. You need to get a daily build dated July 16 or later. I tried opening the sequence view for 1JFF in one of the successful daily builds and it was correct, e.g.:
You can see that the interior gap is correctly recognized now. Hopefully the daily builds on the Mac platforms will work tonight.
thanks for your suggestions. Unfortunately removing the SEQRES record did not help. I also downloaded and installed the new chimera-version today. The false numbering in the alignment still occurs. Also when I remove the SEQRES record and try with the new chimera version the false numbering appears. Quite annoying... The pdb-file is not in the protein data bank yet.
Ah, the "remove SEQRES records" hack will only work for the single- sequence viewer, not an alignment. I tested that in the 1.4.1 release and it indeed worked as I expected:
The whole sequence is smushed together with no gaps, but the correspondences are correct. You could then align additional sequences to the smushed sequence and the correspondences would remain correct. Nonetheless it's probably simply better to wait for a daily build to succeed than to use the hack now. So make sure you check the date in the download table. If you still have problems with a release later than July 16, let me know.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
On Jul 16, 2010, at 3:06 AM, Gesa Volkers wrote:
Additionally I downloaded 1JFF and did an alignment with another tubulin-structure. And still the numbering does not recognize this gap between 34 and 61.
Gesa Volkers schrieb:
Dear Eric,
thanks for your suggestions. Unfortunately removing the SEQRES record did not help. I also downloaded and installed the new chimera-version today. The false numbering in the alignment still occurs. Also when I remove the SEQRES record and try with the new chimera version the false numbering appears. Quite annoying... The pdb-file is not in the protein data bank yet.
Any more suggestions? Could I add "dummy atoms" in the structure?
Thanks, Gesa
Eric Pettersen schrieb:
On Jul 14, 2010, at 6:53 AM, Gesa Volkers wrote:
Dear All,
I observed an issue with the alignment function of Chimera. I loaded a crystal structure where in between a loop is not modelled due to low electron density- so these residues are missing in the structure and the numbering also has this gap. When I align this structure with others, the missing residues are not recognized and chimera just overrides the numbering so after this loop all the numbers in the alignment do not fit anymore to the residues they should belong to although the structure and numberings are shown in the right way in the graphics window. How can I overcome this issue? Should I change the pdb-file? In which way?
Thanks for all responses!
Hi Gesa, I had a similar problem reported with 1JFF 12 days ago and haven't yet worked on fixing it. It seems to only happen for a small subset of structures. 1JFF has a missing loop and also exactly one missing residue at the N terminus. Is your structure like that? Is it a standard PDB entry (if so, what's the ID code)? I'll add you to the recipient list for the ticket we have open in our bug database for this (#8585) so you will be notified when it gets fixed. It'll probably be fairly soon. Until then the only workaround that occurs to me is for you to remove the SEQRES records from the PDB file before you open it. Then the Sequence viewer will depict only the residues physically present in the structure -- no gaps (incorrect or otherwise) will be depicted.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu-- Gesa Volkers Institut für Biochemie, Molekulare Strukturbiologie Felix-Hausdorff-Straße 4 17489 Greifswald Germany +49 (3834)-864392
This message and the one before it reappeared due to a malfunction in our mailing-list software. That problem has been fixed. Sorry for the spam. The problem in the mail itself has been fixed also, as per the response I sent to the original messages.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu
On Jul 16, 2010, at 2:58 AM, Gesa Volkers wrote:
Dear Eric,
thanks for your suggestions. Unfortunately removing the SEQRES record did not help. I also downloaded and installed the new chimera-version today. The false numbering in the alignment still occurs. Also when I remove the SEQRES record and try with the new chimera version the false numbering appears. Quite annoying... The pdb-file is not in the protein data bank yet.
Any more suggestions? Could I add "dummy atoms" in the structure?
Thanks, Gesa
Eric Pettersen schrieb:
On Jul 14, 2010, at 6:53 AM, Gesa Volkers wrote:
Dear All,
I observed an issue with the alignment function of Chimera. I loaded a crystal structure where in between a loop is not modelled due to low electron density- so these residues are missing in the structure and the numbering also has this gap. When I align this structure with others, the missing residues are not recognized and chimera just overrides the numbering so after this loop all the numbers in the alignment do not fit anymore to the residues they should belong to although the structure and numberings are shown in the right way in the graphics window. How can I overcome this issue? Should I change the pdb-file? In which way?
Thanks for all responses!
Hi Gesa, I had a similar problem reported with 1JFF 12 days ago and haven't yet worked on fixing it. It seems to only happen for a small subset of structures. 1JFF has a missing loop and also exactly one missing residue at the N terminus. Is your structure like that? Is it a standard PDB entry (if so, what's the ID code)? I'll add you to the recipient list for the ticket we have open in our bug database for this (#8585) so you will be notified when it gets fixed. It'll probably be fairly soon. Until then the only workaround that occurs to me is for you to remove the SEQRES records from the PDB file before you open it. Then the Sequence viewer will depict only the residues physically present in the structure -- no gaps (incorrect or otherwise) will be depicted.
--Eric
Eric Pettersen UCSF Computer Graphics Lab http://www.cgl.ucsf.edu-- Gesa Volkers Institut für Biochemie, Molekulare Strukturbiologie Felix-Hausdorff-Straße 4 17489 Greifswald Germany +49 (3834)-864392
Chimera-dev mailing list Chimera-dev@cgl.ucsf.edu http://plato.cgl.ucsf.edu/mailman/listinfo/chimera-dev
participants (2)
-
Eric Pettersen -
Gesa Volkers