You are here

How do you implement phosphoserine in ab -initio folding?

11 posts / 0 new
Last post
How do you implement phosphoserine in ab -initio folding?
#1

I would like to substitute one of the serines (S43) in my sequence to be phosphorylated. How would I do that?

I tried substituting [SER_p] in the fasta sequence and -residues:patch_selectors phosphorylation but that did not work. Any suggestions?

Thanks!

Post Situation: 
Fri, 2011-12-16 14:45
smiruthi

I think you read SER_p from annotated sequence files that look like [SER_p:Cterminusful] or something like that. The problem is that the p does not stand for phosphate, it stands for patch, indicating it is a patched residue.

I don't know what the annotation code for phosphoserine is. I think it will be [SER_p:phosphorylated], assuming it's otherwise a normal (not terminus, etc) residue. That's from its patch file at rosetta_database/chemical/residue_type_sets/fa_standard/patches.

To get it for sure, find (any) phosphoserine PDB. Get Rosetta to read it in correctly with score_jd2, using the -out:pdb flag, and make sure the output PDB still has phosphoserine. If that works, then use -out:file:silent result.out -out:file:silent_struct_type binary on the same file; the result silent file will have an annotated sequence line. You can copy the phosphoserine residue name from that. Let me repeat: this doesn't have to be your protein, just any P-ser to copy from.

Sun, 2011-12-18 16:00
smlewis

I tried [SER_p:phosphorylated] (and also confirmed the usage using score_jd2 with a phosphoserine pdb) and still obtained the error:

ERROR: unrecognized residue name 'SER_p:phosphorylated'
ERROR:: Exit from: src/core/chemical/ResidueTypeSet.hh line: 151

Is there anything else I can try?

Mon, 2011-12-19 14:51
smiruthi

Er, I should have checked this first - is it available as a residue type? Try checking patches.txt (the full path to approximately there was in my earlier post) to see that it's not commented out (no # in front of it). Also try running Rosetta in super-verbose mode (I think it's -run:level 500; I know there's a thread about it on the board somewhere). It will output a huge list of all the residue types that exist, look for your phosphoserine there to confirm Rosetta is creating it properly.

I can find that flag for you later if you can't find it.

Mon, 2011-12-19 15:53
smlewis

I think I might know what the problem is. The centroid folding aspect of ab initio didn't recognize the Ser_p:phosphorylated patch. I tried -ignore_unrecognized_res but it didn't work, so I added a ser_phosphorylated.txt (similar to tyr_sulphated patch) modified centroid/patches.txt to include the same.

Now I have a different problem:
.
.
.
===================================================================
Finished Abinitio
.
.
.
can not find a residue type that matches the residue SER_p:phosphorylatedat position 43

ERROR: core::util::switch_to_residue_type_set fails

ERROR:: Exit from: src/core/util/SwitchResidueTypeSet.cc line: 135

Any suggestions?
To answer your previous question the patches.txt (../fa_standard/) file does contain the uncommented path to the ser_phosphorylated.txt and -run:level 500 doesn't work.

Mon, 2011-12-19 16:46
smiruthi

The [SER_p:phosphorylated] patch works separately but it doesn't work in conjugation with another patch. i.e. can't use two patches together. The centroid fix works well for ab initio folding.

Tue, 2011-12-20 18:25
smiruthi

I know this is an old thread, but I have the exact same problem. My patch file appears to have worked for ab-initio, but not in conjunction with another patch. Any advice?

Mon, 2012-03-19 13:47
brspurri

Can you be more specific? Does it not work because the double-patched residue type is not created, or because the double-patched residue type is dysfunctional? You can check what gets created with -out:level 500, which will produce *gobs* of output on which residue types exist.

Mon, 2012-03-19 13:53
smlewis

My specific error is:
can not find a residue type that matches the residue ASN_p:glyat position 70
ERROR: core::util::switch_to_residue_type_set fails
ERROR:: Exit from: src/core/util/SwitchResidueTypeSet.cc line: 135

Applying -out:levels does give a ton of information. What am I looking for in it though?

The score application reads the fullatom patch just fine.

Mon, 2012-03-19 14:27
brspurri

That error message looks to me like you're having a centroid/fullatom issue. You probably should check that the residue patches you're trying to use exist both in the centroid parameter set and the full atom parameter set. (Also, you're not using the -correct flag, by any chance, are you?)

Regarding the output, usually the relevant lines are the ones at the very end - the ones which are output just before the program terminates. - But Steven is also referring to early on in the log where you'll see a large number of lines like

core.chemical: successfully patched: SER_p:NtermProteinFull to: SER_p:NtermProteinFull_p:phosphorylated

You can check those lines to see if your desired patched residue was actually created.

Mon, 2012-03-19 17:22
rmoretti

Looks like that was it. I needed to include the patch file in the centroid directory as well. Thanks everybody.

Tue, 2012-03-20 07:52
brspurri