rosetta_scripts_BOGUS coords warning message

[Anonymous]

 

 Hi,

I am using rosetta scripts to build a protein into electron density.

However, I am getting some warning messages, which relate to my CYX residues. I understand the message itself, but I don’t know why it appears, what it means to my structure and whether I should be worried about it. As I said, I have several CYS residues, which I mutated to CYX (CYX name in PDB file) and when I start my calculations I am getting the following warning message for all CYX residues:

core.chemical.AtomICoor: Warning from IcoorAtomID::xyz(): ICoorAtomID xyz depends on invalid residue connection, returning BOGUS coords (null vector): this_rsd= CYX 352 connid= 3 partner_seqpos= 0

core.chemical.AtomICoor: Warning from IcoorAtomID::xyz(): ICoorAtomID xyz depends on invalid residue connection, returning BOGUS coords (null vector): this_rsd= CYX 355 connid= 3 partner_seqpos= 0

… and so on

core.conformation.Conformation: Failed to find a residue connection for residue 352 with connection point 3

core.conformation.Conformation: Connecting residues: 355 ( CYX ) and 399 ( CYX ) at atoms  SG  and  SG 

… and so on

 

The weirdest thing is that this message appears only when I submit my job in a threaded mode. When I run it on a single core, I don’t have this kind of problem.

I am using Rosetta version: rosetta_bin_linux_2016.32.58837_bundle

Flags below:

rosetta_scripts.linuxgccrelease 

-database database/ 

-in::file::PDB.pdb 

-parser::protocol xml.xml 

-auto_setup_metals 

-detect_disulf false 

-extra_res_fa CYX.params 

-relax:min_type lbfgs_armijo_nonmonotone 

-default_max_cycles 200 

-relax:jump_move true 

-edensity:mapreso X 

-edensity::cryoem_scatterers 

-crystal_refine 

 

Any comments will be greatly appreciated

Thanks,

Karl 

[Anonymous]

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX?  

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.  Make a deep copy of your install and recompile both?  (deep copy in case the “old verison” was good, and we don’t want to remove it!).  This is especially true if the database you have matches to one version but not the other.

If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do.

 

 

[Anonymous]

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX?  

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.  Make a deep copy of your install and recompile both?  (deep copy in case the “old verison” was good, and we don’t want to remove it!).  This is especially true if the database you have matches to one version but not the other.

If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do.

 

 

[Anonymous]

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX?  

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.  Make a deep copy of your install and recompile both?  (deep copy in case the “old verison” was good, and we don’t want to remove it!).  This is especially true if the database you have matches to one version but not the other.

If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do.

 

 

[Anonymous]

hey,

thank you for your reply.

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX? 

Answer: I wany CYX. These should bind to Fe atoms of the iron-sulfr clusters.

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.

No. I compiled only one version of the code and I use it at the login node (without CYX warning message) and in a threaded mode = each core does a separate job, i.e, 20 cores = 20 different processes with different random seed, and then I see this message.

 If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do

I tghout that I don’t have to do it anymore. Thats way you (Rosetta crew) have introduced -auto_setup_metals function and thats way I used CYX, which stores information about connectivity between sulfide atom and metal ion. Right? Correct me if I am wrong.

If I am wrong I would be grateful if you can send me an example of a file, showing how to define a link record between CYS and a desired metal ion.

thanks

karl 

 

[Anonymous]

hey,

thank you for your reply.

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX? 

Answer: I wany CYX. These should bind to Fe atoms of the iron-sulfr clusters.

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.

No. I compiled only one version of the code and I use it at the login node (without CYX warning message) and in a threaded mode = each core does a separate job, i.e, 20 cores = 20 different processes with different random seed, and then I see this message.

 If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do

I tghout that I don’t have to do it anymore. Thats way you (Rosetta crew) have introduced -auto_setup_metals function and thats way I used CYX, which stores information about connectivity between sulfide atom and metal ion. Right? Correct me if I am wrong.

If I am wrong I would be grateful if you can send me an example of a file, showing how to define a link record between CYS and a desired metal ion.

thanks

karl 

 

[Anonymous]

hey,

thank you for your reply.

Big question A) why are you using CYX?  Do you want disulfides, which are CYD, or do you want “cysteine connected to something in an unusual way”, which is CYX? 

Answer: I wany CYX. These should bind to Fe atoms of the iron-sulfr clusters.

if there is different behavior between single and MPI modes (not multithreaded – we don’t support multithreading yet), probably they are compiled from different versions of the code.

No. I compiled only one version of the code and I use it at the login node (without CYX warning message) and in a threaded mode = each core does a separate job, i.e, 20 cores = 20 different processes with different random seed, and then I see this message.

 If you actually want CYX, you probably have to define LINK records, etc, to tell Rosetta what you expected it to connect to.  I can ping for help if that’s what you meant to do

I tghout that I don’t have to do it anymore. Thats way you (Rosetta crew) have introduced -auto_setup_metals function and thats way I used CYX, which stores information about connectivity between sulfide atom and metal ion. Right? Correct me if I am wrong.

If I am wrong I would be grateful if you can send me an example of a file, showing how to define a link record between CYS and a desired metal ion.

thanks

karl 

 

[Anonymous]

You might not need LINK records for this particular problem – auto_setup_metals might handle it.  That said, it’s auto_setup_metals, not auto_setup_iron_sulfur_clusters – I think it does only monatomic metals.  

 

This is enough more information to pass it along to a few devs likely to know something more.  I don’t have any example files but they might.

[Anonymous]

You might not need LINK records for this particular problem – auto_setup_metals might handle it.  That said, it’s auto_setup_metals, not auto_setup_iron_sulfur_clusters – I think it does only monatomic metals.  

 

This is enough more information to pass it along to a few devs likely to know something more.  I don’t have any example files but they might.

[Anonymous]

You might not need LINK records for this particular problem – auto_setup_metals might handle it.  That said, it’s auto_setup_metals, not auto_setup_iron_sulfur_clusters – I think it does only monatomic metals.  

 

This is enough more information to pass it along to a few devs likely to know something more.  I don’t have any example files but they might.

[Anonymous]

So the devs I’ve emailed this too have wandered off into a conversation of how they’d like this system to work…which implies to me that they don’t think it works as is.  I’ll update you if I get something concrete.

[Anonymous]

So the devs I’ve emailed this too have wandered off into a conversation of how they’d like this system to work…which implies to me that they don’t think it works as is.  I’ll update you if I get something concrete.

[Anonymous]

So the devs I’ve emailed this too have wandered off into a conversation of how they’d like this system to work…which implies to me that they don’t think it works as is.  I’ll update you if I get something concrete.

[Anonymous]

thanks. Your help will be greatly appreciated.

[Anonymous]

thanks. Your help will be greatly appreciated.

[Anonymous]

thanks. Your help will be greatly appreciated.

[Author]

Posts