Quick Question on Automodel
Hello all,
I'm trying to model from a known template (ident 37%). The template is missing the first 3 residues in the structure.
- I start reading the template at position 4. - I add 3 dashes to the beginning of the template to compensate for the missing residues. (otherwise I get a "residue mismatch" error from modeller)
Now, my alignment needs 3 additional gaps at the start of the template.
No matter what I do, adding 6 dashes to the beginning of the template (3 for the missing residues and 3 extra for my alignment) has no effect on the outcome and the model doesn't align with the template in the way I expect.
Here is a bit of my ali file. I want the TRL of the template to align with the TRM of the sequence.
When it finishes, the position of TRM has a different offset in relation to TRL Am i doing something wrong here?
>P1;template structureX:template:4:A:330:A:xxxxx:xxxxxxx:2.30:0.211 ------TR*L*......................rest of seq*
>P1;seq sequence:seq:.:.:.:.:::: EAARRRTR*M*........ rest of seq*
Thanks in advance. Pedro
On 2/28/21 1:00 PM, Pedro Guillem wrote: > I'm trying to model from a known template (ident 37%). > The template is missing the first 3 residues in the structure. > > - I start reading the template at position 4. > - I add 3 dashes to the beginning of the template to compensate for the > missing residues. (otherwise I get a "residue mismatch" error from modeller)
Dashes/gaps do not have any effect on which residues are read from the PDB file; this is controlled entirely by the PIR header (the line starting with "structureX", where you - apparently correctly - have 4 as the start residue number). Dashes/gaps only control the correspondence between target and template residues.
> Here is a bit of my ali file. I want the TRL of the template to align > with the TRM of the sequence.
Your alignment looks fine to me, assuming by "..." you mean "the rest of the sequence goes here". (A literal "." has a special meaning to Modeller; read in the manual about modeling with ligands.)
> When it finishes, the position of TRM has a different offset in relation > to TRL
Not sure what you mean here, unless you were expecting T to be numbered 4. Modeller numbers residues in your model starting from 1, regardless of your alignment. If you want a different numbering, use rename_segments(); see https://salilab.org/modeller/10.0/manual/node23.html and https://salilab.org/modeller/10.0/manual/node30.html
Ben Webb, Modeller Caretaker
Thank you Ben.
Effectively. T should be 7 but it is being numbered as 4. Even with dashes indicating 6 gaps before the T.
Sorry if I wasn't clear. By offset I ment the output positions not following my alignment.
Yup. the "..." in the sequence is just the rest of the sequence (to make the message shorter) :)
Best Pedro
On Mon, Mar 1, 2021, 7:42 PM Modeller Caretaker modeller-care@salilab.org wrote:
> On 2/28/21 1:00 PM, Pedro Guillem wrote: > > I'm trying to model from a known template (ident 37%). > > The template is missing the first 3 residues in the structure. > > > > - I start reading the template at position 4. > > - I add 3 dashes to the beginning of the template to compensate for the > > missing residues. (otherwise I get a "residue mismatch" error from > modeller) > > Dashes/gaps do not have any effect on which residues are read from the > PDB file; this is controlled entirely by the PIR header (the line > starting with "structureX", where you - apparently correctly - have 4 as > the start residue number). Dashes/gaps only control the correspondence > between target and template residues. > > > Here is a bit of my ali file. I want the TRL of the template to align > > with the TRM of the sequence. > > Your alignment looks fine to me, assuming by "..." you mean "the rest of > the sequence goes here". (A literal "." has a special meaning to > Modeller; read in the manual about modeling with ligands.) > > > When it finishes, the position of TRM has a different offset in relation > > to TRL > > Not sure what you mean here, unless you were expecting T to be numbered > 4. Modeller numbers residues in your model starting from 1, regardless > of your alignment. If you want a different numbering, use > rename_segments(); see > https://salilab.org/modeller/10.0/manual/node23.html > and > https://salilab.org/modeller/10.0/manual/node30.html > > Ben Webb, Modeller Caretaker > -- > modeller-care@salilab.org https://salilab.org/modeller/ > Modeller mail list: https://salilab.org/mailman/listinfo/modeller_usage >
Hello all,
Any additional comments or suggestions on this issue? So far I have not been able to make modeller follow my alignment.
Is there an *explicit way *to tell modeller which residue (in the template) must be aligned with which residue in the sequence?
ie: align->template:1:21 to sequence:7:27
I'm looking to keep the structure identical in some segments.. and have modeller make a best guess in other parts.
Best Pedro
El lun, 1 mar 2021 a las 20:01, Pedro Guillem (pedro.guillem@gmail.com) escribió:
> Thank you Ben. > > Effectively. T should be 7 but it is being numbered as 4. Even with dashes > indicating 6 gaps before the T. > > Sorry if I wasn't clear. By offset I ment the output positions not > following my alignment. > > Yup. the "..." in the sequence is just the rest of the sequence (to make > the message shorter) :) > > Best > Pedro > > > On Mon, Mar 1, 2021, 7:42 PM Modeller Caretaker modeller-care@salilab.org > wrote: > >> On 2/28/21 1:00 PM, Pedro Guillem wrote: >> > I'm trying to model from a known template (ident 37%). >> > The template is missing the first 3 residues in the structure. >> > >> > - I start reading the template at position 4. >> > - I add 3 dashes to the beginning of the template to compensate for the >> > missing residues. (otherwise I get a "residue mismatch" error from >> modeller) >> >> Dashes/gaps do not have any effect on which residues are read from the >> PDB file; this is controlled entirely by the PIR header (the line >> starting with "structureX", where you - apparently correctly - have 4 as >> the start residue number). Dashes/gaps only control the correspondence >> between target and template residues. >> >> > Here is a bit of my ali file. I want the TRL of the template to align >> > with the TRM of the sequence. >> >> Your alignment looks fine to me, assuming by "..." you mean "the rest of >> the sequence goes here". (A literal "." has a special meaning to >> Modeller; read in the manual about modeling with ligands.) >> >> > When it finishes, the position of TRM has a different offset in >> relation >> > to TRL >> >> Not sure what you mean here, unless you were expecting T to be numbered >> 4. Modeller numbers residues in your model starting from 1, regardless >> of your alignment. If you want a different numbering, use >> rename_segments(); see >> https://salilab.org/modeller/10.0/manual/node23.html >> and >> https://salilab.org/modeller/10.0/manual/node30.html >> >> Ben Webb, Modeller Caretaker >> -- >> modeller-care@salilab.org https://salilab.org/modeller/ >> Modeller mail list: https://salilab.org/mailman/listinfo/modeller_usage >> >
On 3/8/21 4:03 AM, Pedro Guillem wrote: > Is there an *explicit way *to tell modeller which residue (in the > template) must be aligned with which residue in the sequence?
Yes, this is what your alignment is for. If you align residue X in your target with residue Y in your template, that's what you'll get in your output model, end of story.
> I'm looking to keep the structure identical in some segments.. and have > modeller make a best guess in other parts.
See https://salilab.org/modeller/10.0/manual/node23.html
Ben Webb, Modeller Caretaker
participants (2)
-
Modeller Caretaker -
Pedro Guillem