span and frags filefor multichain transmembrane protein

[Anonymous]

Dear All:

 

I’m quite new to Rosetta and I current got a problem when try to modelling a two-domain transmembrane protein.

 

The protein has two transmembrane domain (chain A, start from residue 1 and chain B, also start from residue 1). There is no covalent linking between the two domains.

 

The template I used is a crystal structure of another protein from same family, which also has two domains and got ~60% sequence similarity. The main different is the protein been modelled has some long loops between transmembrane helix whereas the template don’t have.

 

As the template contains the information of surface interactions between domains, I think I should model them together. I’m following the tutorial in https://www.rosettacommons.org/demos/latest/tutorials/rosetta_cm/rosetta_cm_tutorial 

 

However, which does not contain information on how make Rosetta read span and frag files for different domain.

 

Thanks a lot in advance!!

 

Guang

 

 

 

[Anonymous]

Hi Guang,

I think the easiest in your case is to treat the second chain as a continuation of the first one. So chain B can remain chain B, but the residue numbering should continue from chain A (and not start at 1). This way, you can just add chain B in the span file, in which the residue numbers are treated as ‘unique identifiers’. Information on how the spanfile format looks like can be found here: https://www.rosettacommons.org/docs/latest/application_documentation/membrane_proteins/RosettaMP-GettingStarted-PreparingInputs

I would do the same for the fragment files, i.e. treat is as a single chain – the termini typically aren’t super trustworthy when modeled from a homologue, so Rosetta will optimize that anyway. 

Hope this helps,

Julia

 

[Anonymous]

Hi Guang,

I think the easiest in your case is to treat the second chain as a continuation of the first one. So chain B can remain chain B, but the residue numbering should continue from chain A (and not start at 1). This way, you can just add chain B in the span file, in which the residue numbers are treated as ‘unique identifiers’. Information on how the spanfile format looks like can be found here: https://www.rosettacommons.org/docs/latest/application_documentation/membrane_proteins/RosettaMP-GettingStarted-PreparingInputs

I would do the same for the fragment files, i.e. treat is as a single chain – the termini typically aren’t super trustworthy when modeled from a homologue, so Rosetta will optimize that anyway. 

Hope this helps,

Julia

 

[Anonymous]

Hi Guang,

I think the easiest in your case is to treat the second chain as a continuation of the first one. So chain B can remain chain B, but the residue numbering should continue from chain A (and not start at 1). This way, you can just add chain B in the span file, in which the residue numbers are treated as ‘unique identifiers’. Information on how the spanfile format looks like can be found here: https://www.rosettacommons.org/docs/latest/application_documentation/membrane_proteins/RosettaMP-GettingStarted-PreparingInputs

I would do the same for the fragment files, i.e. treat is as a single chain – the termini typically aren’t super trustworthy when modeled from a homologue, so Rosetta will optimize that anyway. 

Hope this helps,

Julia

 

[Anonymous]

Thank you Julia!

As in my case the terminals are far away in space (in the opposite side of the interface), may be the only reasonable way is build long glycine chain connecting the domains? (for both the template and the target )?

[Anonymous]

Thank you Julia!

As in my case the terminals are far away in space (in the opposite side of the interface), may be the only reasonable way is build long glycine chain connecting the domains? (for both the template and the target )?

[Anonymous]

Thank you Julia!

As in my case the terminals are far away in space (in the opposite side of the interface), may be the only reasonable way is build long glycine chain connecting the domains? (for both the template and the target )?

[Anonymous]

Depending on the modeling protocol, you shouldn’t need to add a linker.

Most Rosetta protocols (including RosettaCM, I belive) are smart enough not to bother making fragment insertions across the chain terminii. So there shouldn’t be any issue with giving it a fragment file (or span file) which treats the two chains as a single chain.

What you will need to be careful of is the format of the fasta file you use with RosettaCM – you’ll need to tell Rosetta that they’re two separate chains, as opposed to a single chain. I belive that’s as easy as adding a slash (‘/’) between the end of the first chain and the start of the second. 

[Anonymous]

Depending on the modeling protocol, you shouldn’t need to add a linker.

Most Rosetta protocols (including RosettaCM, I belive) are smart enough not to bother making fragment insertions across the chain terminii. So there shouldn’t be any issue with giving it a fragment file (or span file) which treats the two chains as a single chain.

What you will need to be careful of is the format of the fasta file you use with RosettaCM – you’ll need to tell Rosetta that they’re two separate chains, as opposed to a single chain. I belive that’s as easy as adding a slash (‘/’) between the end of the first chain and the start of the second. 

[Anonymous]

Depending on the modeling protocol, you shouldn’t need to add a linker.

Most Rosetta protocols (including RosettaCM, I belive) are smart enough not to bother making fragment insertions across the chain terminii. So there shouldn’t be any issue with giving it a fragment file (or span file) which treats the two chains as a single chain.

What you will need to be careful of is the format of the fasta file you use with RosettaCM – you’ll need to tell Rosetta that they’re two separate chains, as opposed to a single chain. I belive that’s as easy as adding a slash (‘/’) between the end of the first chain and the start of the second. 

[Anonymous]

Thank you! That solve my puzzle, I misunderstand that I must make chain marker in the alignment file.

[Anonymous]

Thank you! That solve my puzzle, I misunderstand that I must make chain marker in the alignment file.

[Anonymous]

Thank you! That solve my puzzle, I misunderstand that I must make chain marker in the alignment file.

[Author]

Posts