You are here

to create the template pdb file

9 posts / 0 new
Last post
to create the template pdb file
#1

Dear everyone.

I used the rosetta membrane homology modeling, recently, when I tried to create the template pdb file by the script createTemplate.pl, the error of "attempt to assign missing density from parent[86] as a takeoff for query[167]" occurred. it righet that the parent pdb file do have 86 residues.

my zone file is :
zone 17-29 : 1-13
zone 41-46 : 14-19
zone 80-82 : 20-22
zone 88-92 : 23-27
zone 110-167 : 29-86

the aligment file is:
>quary
SVKKLADVMVLTVFCLSVFALIGLQLFMGNLRHKCVRNFTALNGTNGSVE
ADGLVWESLDLYLSDPENYLLKNGTSDVLLCGNSSDAGTCPEGYRCLKAG
ENPDHGYTSFDSFAWAFLALFRLMTQDCWERLYQQTLRSAGKIYMIFFML
VIFLGSFYLVNLILAVVAMA

>model
----------------SVAALLTVVFYIA-----------AVMATN----
-----------------------------LYG-----ATFPE-W------
---------FGDLSKSLYTLFQVMTLESWSMGIVRPVMNVHPNAWVFFIP
FIMLTTFTVLNLFIGII---

please help me!

Thanks and best regards!

xiaofeng

Post Situation: 
Sat, 2012-12-29 23:04
benniu2004

I'm guessing that the occupancy (see http://bmerc-www.bu.edu/needle-doc/latest/atom-format.html) of the Calpha atom of one of the residues in your template PDB file is set to zero.

In general, if you want Rosetta or Rosetta-related tools to "count" residues/atoms in a PDB, they need to have non-zero occupancies (preferably equal to 1.00).

Tue, 2013-01-01 11:55
rmoretti

Thank you very much for your reply.

Maybe you are right, but when checked the parent PDB file, I found none of residues was set to zero.

The template file has 86 residues is right and the zone file showed that the parent file should be 86 residues. I puzzled why this appeared 167.

Tue, 2013-01-01 16:40
benniu2004

Your query sequence (the first one) in your alignment is the one that has 167 residues. (It actually has 170, but it's residue 167 that's being aligned to residue 86 of your model file.)

The error message you're getting really looks to be an issue with occupancy. Could you post the parent PDB file, so I could take a look at it? (You may need to change the extension to ".txt" to get the forum software to let you attach it.) Also, could you post the command line you used to run the program?

Wed, 2013-01-02 10:48
rmoretti

thank you for your attention~~

The parent pdb file was attached in the attachment. please check,thanks!

the command line I have used is: /scripts/fastaAln2zones.pl ${alignment}/domain1.clustalw domain1.zone

The .clustalw and .zone files were just as pasted top.

Wed, 2013-01-02 22:00
benniu2004

Rocco is right that the template PDB is subtly corrupted. All of the columns after the residue number are shifted one column to the left. I'm not positive this is the source of your errors but it's a good guess. (we can also try adding a chain ID to the PDB; I'm not sure that that matters).

A corrected PDB is attached.

Thu, 2013-01-03 07:30
smlewis

Yes, just as you say "All of the columns after the residue number are shifted one column to the left" , but when I used this corrected file as the parent pdb file it also give me the alert: attempt to assign missing density from parent[86] as a takeoff for query[167]. what a crazy thing!!

In addition, the parent file was the output file of script:renum.awk, I have checked my template file, it is a standard pdb file, all of the columns are in the correct position. what has happened?

Even more unbelievable is that I have done some work by rosetta_membrane_homology_modeling and this is no errors.

Maybe distance gap between the target and the template is great? (170:86)

Who will help me! please! Thanks!!

Thu, 2013-01-03 19:14
benniu2004

I tried running the createTemplate.pl script with Steven's adjusted template file with "perl createTemplate.pl -parentpdb template.txt -zonesfile zones.txt -fastafile align.txt -outpdb out.pdb", (I'm unable to find the fastaAln2zones.pl script) and am getting an error message, although it's not the same error message you said you got previously:

createTemplate.pl: attempt to exceed boundaries of parent for takeoff for query residue 15 at parent residue -1. recommend trimming back alignment by at least one residue

This can be simply fixed by adding "-takeoffpad F" to the command line, which results in a command that completes without errors. (https://www.rosettacommons.org/guide/Comparative+Modeling indicates that trimming back the zones by one residue is recommended - I think the option controls this automatically.)

Fri, 2013-01-04 12:21
rmoretti

Thank you very much rmoretti, when I adding "-takeoffpad F" to the command line, the script can run. there is no error appeared again.

Best regards!

Fri, 2013-01-04 17:20
benniu2004