[Date Prev][Date Next][Thread Prev][Thread Next][Date Index][Thread Index]

[IMP-users] Re: contraints



On 7/3/25 2:25 AM, Michael Brunsteiner via IMP-users wrote:
> regarding the antibody/IGG modelling, one more question: I
> understand, with modeller, you can define S-S bonds, and modelling
> the FC-part plus hinge (including di-sulfide bonds) does work if you
> have a decent template for the FC (without hinge) - of which there
> are many. I tried this and it does give the expected result. To get
> the full length antibody, this would require including the two Fab
> domains, ideally as rigid bodies. I understand rigid bodies can be
> defined in modeller, but I wonder: is there a way to define these
> two Fab domains, and perhaps also the FC,  as rigid bodies, have
> their initial structures be random positions/orientations, and then
> ask modeller to construct a model of the three domains (one FC, two
> Fabs) with only the hinge (including the S-S constraints) being
> sampled, with the constraint that there are supposed to be no
> overlaps/clashes between the three domains?

Modeller does support rigid bodies, but the support is somewhat 
rudimentary and isn't employed by any of the commonly-used protocols. 
You could certainly achieve this though by building a regular 
comparative model aligned to multiple templates. Each part of your model 
aligned to a template will be constrained to resemble that template, so 
will be close to rigid (particularly if your sequence identity is high). 
But restraints are not constructed *between* templates, so hinge motion 
would be allowed. Your alignment might look something like

Fab1  XXXXX/----/-----*
Hinge -----/XXXX/-----*
Fab2  -----/----/XXXXX*
Model YYYYY/YYYY/YYYYY*

	Ben
-- 
                      https://salilab.org/~ben/
"It is a capital mistake to theorize before one has data."
	- Sir Arthur Conan Doyle