Creating centroid patches for Proline (pro_hydroxylated_case1 and pro_hydroxylated_case2)

[Anonymous]

Greetings,

When trying to dock two proteins (with ligands), I obtained the following error:

can not find a residue type that matches the residue PRO_p:pro_hydroxylated_case1at position 31

ERROR: core::util::switch_to_residue_type_set fails

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

When I review the log file, I see the following:

protocols.jd2.PDBJobInputter: PDBJobInputter::pose_from_job
core.chemical.ResidueTypeSet: Finished initializing fa_standard residue type set. Created 4219 residue types
core.pack.task: Packer task: initialize from command line()
protocols.jd2.PDBJobInputter: filling pose from PDB ../FMOandCsmA_J.pdb
protocols.docking.DockingProtocol: Setting docking foldtree
protocols.docking.DockingProtocol: old fold tree: FOLD_TREE EDGE 1 358 -1 EDGE 1 359 1 EDGE 1 360 2 EDGE 1 361 3 EDGE 1 362 4 EDGE 1 363 5 EDGE 1 364 6 EDGE 1 365 7 EDGE 1 366 8 EDGE 1 367 9 EDGE 367 724 -1 EDGE 1 725 10 EDGE 1 726 11 EDGE 1 727 12 EDGE 1 728 13 EDGE 1 729 14 EDGE 1 730 15 EDGE 1 731 16 EDGE 1 732 17 EDGE 1 733 18 EDGE 733 1090 -1 EDGE 1 1091 19 EDGE 1 1092 20 EDGE 1 1093 21 EDGE 1 1094 22 EDGE 1 1095 23 EDGE 1 1096 24 EDGE 1 1097 25 EDGE 1 1098 26 EDGE 1 1099 27 EDGE 1099 1157 -1 EDGE 1 1158 28
protocols.docking.DockingProtocol:
protocols.docking.DockingProtocol: new fold tree: FOLD_TREE EDGE 1 358 -1 EDGE 358 359 6 EDGE 359 366 -1 EDGE 366 367 5 EDGE 367 553 -1 EDGE 553 1158 1 EDGE 553 724 -1 EDGE 724 725 4 EDGE 725 732 -1 EDGE 732 733 3 EDGE 733 1090 -1 EDGE 1090 1091 2 EDGE 1091 1098 -1 EDGE 1158 1157 7 EDGE 1157 1099 -1
protocols.docking.DockingProtocol:
protocols.docking.DockingProtocol: Danger Will Robinson! Native is an impostor!
protocols.docking.DockingProtocol: ////////////////////////////////////////////////////////////////////////////////
protocols.docking.DockingProtocol: /// Rosetta 3 Docking Protocol ///
protocols.docking.DockingProtocol: /// ///
protocols.docking.DockingProtocol: /// Dockable Jumps: 1 ///
protocols.docking.DockingProtocol: /// Low Resolution Docking Protocol: on ///
protocols.docking.DockingProtocol: /// High Resolution Docking Protocol: on ///
protocols.docking.DockingProtocol: /// Low Resolution Filter: on ///
protocols.docking.DockingProtocol: /// High Resolution Filter: on ///
protocols.docking.DockingProtocol: ////////////////////////////////////////////////////////////////////////////////
core.chemical.ResidueTypeSet: Finished initializing centroid residue type set. Created 1203 residue types

So it seems to have something to do with the centroid set of residues.

Fortunately, the cause of this kind of error was nicely explained in: http://www.rosettacommons.org/content/how-do-you-implement-phosphoserine-ab-initio-folding
Namely, it seems that there is no centroid equivalent of the pro_hydroxylated_case1 patch

And indeed, when I comment out those two patches from the full-atom patches.txt by using a # at the beginning of the line, I obtain the following:
core.chemical.ResidueTypeSet: Finished initializing fa_standard residue type set. Created 4005 residue types
core.io.pdb.file_data: [ WARNING ] discarding 2 atoms at position 31 in file ../FMOandCsmA_J.pdb. Best match rsd_type: PRO
…
(repeats similar message a bunch of times)
…
core.io.pdb.file_data: [ WARNING ] can’t find atom for res 31 atom OD1 (trying to set temp)
…
(repeats similar message a bunch of times)
…
And then docking_protocol goes ahead and runs, so I seem to have localized the issue to the absence of the pro_hydroxylated patches for centroid mode.

So here is the question: given the full-atom patches for pro_hydroxylated_case1 and pro_hydroxylated_case2 (which I have attached below), how does one go about making the corresponding centroid patches?
Is there a semi-automated process, or do I need to go through each full-atom patch file line by line deleting hydrogen atoms and references to hydrogen atoms to obtain their centroid equivalents?

Is Andy Chen (the author of the original patch) available for comment? Any advice and or assistance would be appreciated!

Thanks,
Sándor

[Anonymous]

At centroid granularity, the existing proline centroid is good enough for use as hydroxyproline centroid. Your patch only needs to change the name. Try using the fullatom patches but with all the guts commented out as centroid patches. If that won’t work, copy the centroid proline params file and change its name (filename and internal contents) to make it pretend to be hydroxyproline.

[Anonymous]

Hi Sandor, Steven,

I think I got the same problem. The centroid patch file pro_hydroxylated_case1 in the patches folder doesn’t seem to work while the fullatom one works fine.
How did you solve the problem? I am using 3.5 for docking.
Thanks,
T

[Anonymous]

I think It is the same with yours before:

can not find a residue type that matches the residue PRO:pro_hydroxylated_case1at position 382

ERROR: core::util::switch_to_residue_type_set fails

ERROR:: Exit from: src/core/util/SwitchResidueTypeSet.cc line: 325
protocols.jd2.JobDistributor:

[ERROR] Exception caught by JobDistributor for job top-cluster-5_0001

[ERROR] EXCN_utility_exit has been thrown from: src/core/util/SwitchResidueTypeSet.cc line: 325
ERROR: core::util::switch_to_residue_type_set fails

[Anonymous]

Hi Rocco,

I did pay attention to that and I can make sure that is not the case here. There must be something else.
Thanks

[Anonymous]

I ran the docking with the flag and it didn’t print out either “Reading …centroid…pro_hydroxylated_case1.txt” or “successfully patched: PRO to: PRO_p:pro_hydroxylated_case1”. Meanwhile, the fa_standard patch file was read just fine.
Yes, the HYP residues are in middle of the chain as normal residues. Any suggestion?

[Anonymous]

Hello all,
I was wondering if a solution was found for the proline patching problem. As can be seen in the following error message, the centroid patches is located but not applied. The hydroxyproline patch files located in centroid/patches/ and patches.txt files are attached.

I would greatly appreciate any help sorting this out.
………….
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/D_AA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/CtermProtein.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/NtermProtein.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/disulfide.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/protein_cutpoint_upper.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/protein_cutpoint_lower.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/VirtualBB.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/protein_centroid_with_HA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/VirtualNterm.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/N_acetylated.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/C_methylamidated.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/RepulsiveOnly_centroid.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/ser_phosphorylated.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/thr_phosphorylated.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/CtermConnect.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/NtermConnect.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/pro_hydroxylated_case1.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/pro_hydroxylated_case2.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/ShoveBB.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/dna/LowerDNA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/dna/UpperDNA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/dna/VirtualDNAPhosphate.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/rna/LowerRNA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/rna/UpperRNA.txt
core.chemical: (0) Reading patch file: /Users/aaj30/rosetta_src_2015.39.58186_bundle/main/database/chemical/residue_type_sets/centroid/patches/nucleic/rna/Virtual_Phosphate.txt
core.chemical.ResidueTypeSet: (0) Finished initializing centroid residue type set. Created 1906 residue types
core.chemical.ResidueTypeSet: (0) Total time to initialize 0.039489 seconds.
core.chemical.ResidueType: (0) ERASING LOWER_CONNECT: 1 lcid: 2
core.chemical: (0) successfully patched: PRO to: PRO:NtermProteinFull
core.chemical.ResidueType: (0) ERASING UPPER_CONNECT: 2 lcid: 1
core.chemical: (0) successfully patched: VAL to: VAL:CtermProteinFull
core.chemical.ResidueType: (0) ERASING LOWER_CONNECT: 1 lcid: 2
core.chemical: (0) successfully patched: GLY to: GLY:NtermProteinFull
protocols.jd2.JobDistributor: (0)

[ERROR] Exception caught by JobDistributor for job 2v53.ppk_0001

[ERROR] EXCN_utility_exit has been thrown from: src/core/util/SwitchResidueTypeSet.cc line: 296
ERROR: core::util::switch_to_residue_type_set fails

protocols.jd2.JobDistributor: (0) 2v53.ppk_0001 reported failure and will NOT retry
protocols.jd2.JobDistributor: (0) no more batches to process…
protocols.jd2.JobDistributor: (0) 1 jobs considered, 1 jobs attempted in 2 seconds
caught exception 1 jobs failed; check output for error messages

[Anonymous]

Hi
I seem to have found a way out of the problem (at least on my console)!! Now the hyp is successfully patched in the centroid mode.

Line #2 of the pro_hydroxylated_case1.txt and pro_hydroxylated_case2.txt files located in /centroid/patches reads “TYPES HYDROXYLATION”. Presumably, this is incorrect. It should be “TYPES HYDROXYLATION1” for pro_hydroxylated_case1.txt and “TYPES HYDROXYLATION2” for pro_hydroxylated_case2.txt.

aaj

[Anonymous]

Thanks!

I will try your first suggestion of modifying the fullatom patches to create centroid patches and adding the corresponding lines to patches.txt in the centroid folder.

[Anonymous]

Hi,

I’ve tried both the suggested approaches.

First, I duplicated the fullatom patch in the centroid patches folder and then commented out everything after “END SELECTOR”–nothing happened.
Second, I duplicating the proline centroid params file and changed the NAME line to match the patch–again nothing happened.

I made sure to add lines to patches.txt and/or residue_types.txt, but neither seems to be working.
Either way I try it, I still get the same error:

can not find a residue type that matches the residue PRO_p:pro_hydroxylated_case1at position 31

ERROR: core::util::switch_to_residue_type_set fails

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

Please advise!

Thanks,
Sándor

[Anonymous]

It may be that your centroid patches are working, but now it’s failing _exiting_ centroid and returning to fullatom? Is pro_hydroxylated_case1 active in the fullatom set as well as the centroid set? Or is it still failing at the centroid step? At one point you tried commenting it out…

[Anonymous]

Sorry, perhaps I can be more clear…

The only way I can get docking_protocol to run is to comment out the fullatom patches, in which case, the “offending” atoms are deleted and new coordinates are guessed based on the fullatom PRO.params file.
This has no trouble switching between fullatom and centroid modes and docking_protocol completes.

If I turn the fullatom pro_hydroxylated_case1 patch on, regardless of what I do to the centroid patches or parameters, it gives me the same error I mentioned above. I suspect this means I haven’t created these patches/parameters correctly.

Returning to your earlier suggestion, then, I would ask for a bit more clarification:
Which part of the “guts” of the fullatom patch file should I try to comment out?
Alternatively, which lines in the centroid PRO.params file should I try modifying?

Thanks,
Sándor

[Anonymous]

Try these. If they work, let me know and I’ll queue them for the next release. If not, maybe send me a toy problem so I can debug it here…

[Anonymous]

I added the patches to the residue_type_sets/centroid/patches/ folder and specified them in residue_type_sets/centroid/patches.txt
When running, I now get to building the centroid residue typeset at which point I obtain the new error:

ERROR: Residue names must be unique!
ERROR:: Exit from: src/core/chemical/ResidueTypeSet.cc line: 374

At least we’re getting a different error!
Commenting out one of the patches (e.g. case2, leaving case1 active) has no effect–the error persists.

Per your request, I’ve attached a toy system which reproduces the problem.
It consists of a single monomer of the larger trimer and the protein to be docked (I’ve added .txt to the file extension to allow it to be uploaded)
The ligands have been omitted for simplicity.

Thanks!
Sándor

[Anonymous]

The “Residue names must be unique!” error was fixed by realizing that I had (at one point) added another centroid residue type, PRO_H, based on PRO without changing the NAME entry from PRO to PRO_H.
The docking_protocol is working for the test case (i.e. able to change between full-atom and centroid modes).
The two patches may be added to future releases and we can mark this as solved!

Thanks Steven!

Sándor

[Anonymous]

What’s the error message you’re getting? The changes Steven suggested should be incorporated into the Rosetta3.5 release.

[Anonymous]

I think I see what the problem might be. Take a look at rosetta_database/chemical/residue_type_sets/centroid/patches.txt In order for the patch file to actually be loaded, there needs to be a line “patches/pro_hydroxylated_case1.txt” in that file. If there isn’t, try adding it and re-running.

[Anonymous]

Can you check if the patch is being applied appropriately?

If you add the option “-out:levels core.chemical:debug” to the commandline, you’ll turn on debugging output for the patching system. This will generate a *lot* of output. But somewhere in there you should see both “Reading patch file: PATH/rosetta_database/chemical/residue_type_sets/fa_standard/patches/pro_hydroxylated_case1.txt” and “Reading patch file: PATH/rosetta_database/chemical/residue_type_sets/centroid/patches/pro_hydroxylated_case1.txt” (the PATH will vary based on where the database is. Soon after both of those lines, you should see a line like “successfully patched: PRO to: PRO_p:pro_hydroxylated_case1”

If you don’t see the “Reading …centroid…pro_hydroxylated_case1.txt”, Rosetta isn’t loading the centroid patch for some reason. (Are you getting other files in that directory loaded?) If you’re not getting the “successfully patched” line, something else is going on.

Just to double check, the hydroxylated proline is in the middle of the chain and isn’t any sort of terminal residue, or next to missing residues, is it? If so, there’s more complicated patching that would likely happen for that residue.

[Anonymous]

Okay, I’m not sure what’s going on, but it does look like Rosetta is not being able to find and load the patches. (It doesn’t look like it’s even looking for them.) Are any of the other centroid patches being read in? (Do they have lines with the “-out:levels core.chemical:debug” option?)

Could you double check that the pro_hydroxylated_case1.txt and pro_hydroxylated_case2.txt files are located in the rosetta_database/chemical/residue_type_sets/centroid/patches/ directory, and that the two files are listed (as “patches/pro_hydroxylated_case1.txt” and “patches/pro_hydroxylated_case2.txt” without any “#”) each on their own line in the rosetta_database/chemical/residue_type_sets/centroid/patches.txt file?

Another option you could try is to copy the two files to your local directory, and then list them directly on the commandline with the -extra_patch_cen flag. (so “-extra_patch_cen pro_hydroxylated_case1.txt pro_hydroxylated_case2.txt”)

[Author]

Posts