Hi,
I'm doing MD simulation study about Glucocorticoid with protein chains. I need to find out missing residues of protein in my PDB file. I already used chimera but didn't find those missing ones. How can I find missing residues?
If residues are not visible in electron density, no coordinates are given in the pdb file. If you need a continuous protein chain for MD, you either need to find an alternative structure in which these missing residues are present, or you need to model the missing parts.
Use Swiss Model, a fully automated protein structure homology-modeling server.
https://swissmodel.expasy.org/
Hi Kalani,
You can find missing residues by Sequence module in Chimera.
Select it like
Tools -> Sequence -> Sequence
and a window showing sequence will be opened. Missing residues are highlighted by RED flames.
Hi Annemarie,
I searched for alternative structure but there are no structures with missing ones. Then the problem is how to model the missing parts?
Hi Yoshinobu,
I tried to find missing ones by sequence module in Chimera.Missing residues were highlighted by RED flames. But I didn't able to repair missing parts using Chimera. Are there any other way to repair missing residues?
Hi,
For modelling missing residues, I recommend homology modelling server SWISS-MODEL, as already suggested by Anu.
https://swissmodel.expasy.org/interactive
Click User Template button, and upload sequence without missing residues and protein_with_missing_residues.pdb as template. The server can model only missing residues well. The atom coordinates of the template protein should little change during modelling.
hello,
I have tried to repair missing residue with https://swissmodel.expasy.org/interactive. their are more than one chain when i upload the second chain it shows "chain should be unique". Any one have idea about this problem. i am not getting this kind of issue.
If you want to use chimera, you also need modeller. You basically do a homology modelling, including modelling missing loops. Here is a recent detailed tutorial on how to do this: https://biochem.wisc.edu/sites/default/files/facilities/bcrf/tutorials/MODELLER/chimera_ii_modeller_gui.pdf
The advantage of going this way is that you can first do the tutorial with the sample files, then do it with your protein and will learn what to look for in the modelling process.
as Yoshinobu suggested, you can also use swissmodel or any other automated homology modelling server - actually you might try both and compare the results to get an idea on how far you can trust such a model.
Hi again,
See the following link, and find "unique":
https://swissmodel.expasy.org/docs/examples
Maybe you have to upload only one "unique" chain with missing residues.
With a multi-chain object, you have to repair each chain separately (You can simply edit the pdb file in a text editor to delete the other chains. You might want to create several loop-inserted model, as some might lead to clashes then you afterwards bring the different models together again to form the full complex (e.g. by superimposing each chain model onto the original structure
REMARK 465 in your pdb file contains the information about missing residues.
If you are using linux:
grep -iE "REMARK 465 "*[A-Z] your_file.pdb | wc -l #will count the missing residues
grep -iE "REMARK 465 "*[A-Z] your_file.pdb #will list the missing residues
- Badges
- Science topic