Dear All, Is it possible to model home-tetramer using modeller? Thank you very much in advance.
Download prohibited? No problem. CHAT from any browser, without download. Go to http://in.webmessenger.yahoo.com/
Manjunatha Karpenahalli wrote: > Is it possible to model home-tetramer using modeller?
Sure, see: http://salilab.org/modeller/9v4/manual/node28.html
That example is for a dimer, but it is straightforward to use a similar script to model a tetramer.
Ben Webb, Modeller Caretaker
Dear Modeller users,
I try to model homedimer with few point mutations using hometrimer as a template. In the alignment file I specified to use only chain A and chain B from the template, but the out put dimer model has different monomer-monomer interface than template. I used standard script( http://salilab.org/modeller/9v4/manual/node28.html) and the alignment file:
>P1;inv3HD sequence:inv3HD:1:.: 331 ::MODELLING PROTEIN:undefined:: MKPVTLYDVAEYAGVSYQTVSRVVNQASHVSAKTREKVEAAMAELNYIPNRVAQQLAGKQSLLIGVATSSLA LHAPSQIVAAIKSRADQLGASVVVPMVERSGVEACKAAVHNLLAQRVSGLIINYPLDDQDAIAVEAACTNVP ALFLDVSDQTPINSIIFSHEDGTRLGVEHLVALGHQQIALLAGPLSSVSARLRLAGWHKYLTLNQIQPIAER EGDWSAMSGFQQTMQMLNEGIVPTAMLVANDQMALGAMRAIAESGLRVGADISVVGYDDTEDSSCYIPPLTT IKQDFRLLGQTSVDRLLQLSQGQAVKGNQLLPVSLVKRKTTLA\MKPVTLYDVAEYAGVSYQTVSRVVNQAS HVSAKTREKVEAAMAELNYIPNRVAQQLAGKQSLLIGVATSSLALHAPSQIVAAIKSRADQLGASVVVPMVE RSGVEACKAAVHNLLAQRVSGLIINYPLDDQDAIAVEAACTNVPALFLDVSDQTPINSIIFSHEDGTRLGVE HLVALGHQQIALLAGPLSSVSARLRLAGWHKYLTLNQIQPIAEREGDWSAMSGFQQTMQMLNEGIVPTAMLV ANDQMALGAMRAIAESGLRVGADISVVGYDDTEDSSCYIPPLTTIKQDFRLLGQTSVDRLLQLSQGQAVKGN QLLPVSLVKRKTTLA* >P1;1EFA structureX:1EFA:2:A: 329 :.:LAC REPRESSOR:undefined:: -KPVTLYDVAEYAGVSYQTVSRVVNQASHVSAKTREKVEAAMAELNYIPNRVAQQLAGKQSLLIGVATSSLA LHAPSQIVAAIKSRADQLGASVVVSMVERSGVEACKTAVHNLLAQRVSGLIINYPLDDQDAIAVEAACTNVP ALFLDVSDQTPINSIIFSHEDGTRLGVEHLVALGHQQIALLAGPLSSVSARLRLAGWHKYLTRNQIQPIAER EGDWSAMSGFQQTMQMLNEGIVPTAMLVANDQMALGAMRAITESGLRVGADISVVGYDDTEDSSCYIPPLTT IKQDFRLLGQTSVDRLLQLSQGQAVKGNQLLPVSLVKRKTTLA* >P1;1EFA structureX:1EFA:2:B: 329 :.:LAC REPRESSOR:undefined:: -KPVTLYDVAEYAGVSYQTVSRVVNQASHVSAKTREKVEAAMAELNYIPNRVAQQLAGKQSLLIGVATSSLA LHAPSQIVAAIKSRADQLGASVVVSMVERSGVEACKTAVHNLLAQRVSGLIINYPLDDQDAIAVEAACTNVP ALFLDVSDQTPINSIIFSHEDGTRLGVEHLVALGHQQIALLAGPLSSVSARLRLAGWHKYLTRNQIQPIAER EGDWSAMSGFQQTMQMLNEGIVPTAMLVANDQMALGAMRAITESGLRVGADISVVGYDDTEDSSCYIPPLTT IKQDFRLLGQTSVDRLLQLSQGQAVKGNQLLPVSLVKRKTTLA*
Thank you very much in advance for any tips
-Manjunatha Karpenahalli
--- On Tue, 29/7/08, Modeller Caretaker modeller-care@salilab.org wrote: From: Modeller Caretaker modeller-care@salilab.org Subject: Re: [modeller_usage] Oligomer Modelling To: manju_88@yahoo.com Cc: modeller_usage@salilab.org Date: Tuesday, 29 July, 2008, 8:28 PM
Manjunatha Karpenahalli wrote: > Is it possible to model home-tetramer using modeller?
Sure, see: http://salilab.org/modeller/9v4/manual/node28.html
That example is for a dimer, but it is straightforward to use a similar script to model a tetramer.
Ben Webb, Modeller Caretaker
Manjunatha Karpenahalli wrote: > I try to model homedimer with few point mutations using hometrimer as a > template. In the alignment file I specified to use only chain A and > chain B from the template, but the out put dimer model has different > monomer-monomer interface than template. I used standard > script( http://salilab.org/modeller/9v4/manual/node28.html) and the > alignment file:
If this is your actual alignment file, then there are some problems with it:
1. The chain break character is a forward slash (/). You seem to be using a backward slash (). That won't work.
2. You don't have anything aligned with your second chain, so Modeller will build a very poor model for that. What you have as your alignment is something like:
aaaa/---- template 1 bbbb/---- template 2 XXXX/YYYY model
The result is that Modeller will use *both* of your templates to model the first chain, but will have no homology information at all for the second - so you'll essentially get an extended chain. If you really want to build a dimer using two single chain templates, your alignment should look something like:
aaaa/---- template 1 ----/bbbb template 2 XXXX/YYYY model
This, however, is still not ideal (see point 3).
3. You have split your template 1EFA into two single chain templates. Since Modeller has no way to tell that these two templates are actually from the same PDB, it can't build distance restraints between the two chains, and thus the interface will not be conserved. In order to conserve the interface, you should use a single dimer template rather than two monomer templates, i.e.
aaaa/bbbb template 1 XXXX/YYYY model
In your case, your alignment file would thus look like:
>P1;inv3HD sequence:inv3HD:1:.: 331 ::MODELLING PROTEIN:undefined:: (sequence of A chain) / (sequence of B chain)
>P1;1EFA structureX:1EFA:2:A: 329 :B:LAC REPRESSOR:undefined:: (sequence of A chain) / (sequence of B chain)
Note that in the header for the 1EFA template you now tell Modeller to read from residue 2 in the A chain all the way through to residue 329 in the B chain. Look at the alignment file that goes with that script I pointed you to (twochain.ali, in examples/automodel/) for a working example. Notice that both the template and the model are dimers.
Ben Webb, Modeller Caretaker
-
Manjunatha Karpenahalli -
Modeller Caretaker