Member Site › Forums › Rosetta 3 › Rosetta 3 – General › Cterm_amidation
- This topic has 1 reply, 2 voices, and was last updated 3 years, 1 month ago by Anonymous.
-
AuthorPosts
-
-
November 3, 2021 at 3:52 pm #3862Anonymous
Hi there,
I am trying to run Flexpepdock for several peptides which have Cterm_amide group. I checked the ATOM TYPES for NH2 which are also correct but while reading pdb it always gives warning:
core.chemical.Patch: [ WARNING ] Patch MethylatedCtermProteinFull implies it can apply to residue type TYR:Cterm_amidation, but actually applying it fails.
core.chemical.Patch: [ WARNING ] You may want to check your patch definitions.
core.chemical.Patch: [ WARNING ] Patch DimethylatedCtermProteinFull implies it can apply to residue type TYR:Cterm_amidation, but actually applying it fails.
core.chemical.Patch: [ WARNING ] You may want to check your patch definitions.
Although, the output pdb still contains Cterm_amide group but the residue is changed to NON_CANONICAL residue and ATOM name to HETATM as below:
HETATM 195 NT TYR B 306 9.915 21.766 -48.540 1.00 0.00 N
HETATM 206 1HN TYR B 306 9.068 21.701 -49.067 1.00 0.00 H
HETATM 207 2HN TYR B 306 10.780 21.900 -49.024 1.00 0.00 H
I want to know what this warning is about and if I should ignore it??
Beside N-terminal cap, my peptide sequence have at different amino-acid positions additional functional groups added, such as methylation, acetylation. Do I need to generate additional paramter files for those residues like mentioned in non-canonical residue papers?
Looking forward for your expert opinion
Mustafa
-
November 6, 2021 at 10:26 pm #16057Anonymous
Hi,
If the modification is present in the output, then the run itself should be fine.
As for the other PTMs: most of the usual PTMs have a patch file already(I personally used phosphorylation and lysine acetylation without a problem). You can look up if any of them is missing, and create one based on their parent residue.
Julia
-
-
AuthorPosts
- You must be logged in to reply to this topic.