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

Re: TRANFER_XYZ (cluster_cut>0)



HElloo,


Obdulia Rabal wrote:
> 
> Hello.
> 
> Thank you for your answer about the modelling loops ( date: 14/02)
> Now,  I have a problem with the different commands.
> 
> I have the main template I want to use to build the model and several
> small templates I only want to use to model the loops.
> My alignment looks like:
> 
> Main template:
> -------Leu-His-Gly-Tyr-Phe-Val-Phe-Gly-Pro-Thr-Gly--------------
> Loop's templates:------___-Ala-Asn-Trp-Tyr-___-Phe-------------------
> Loop's templates:---------------------------- Phe-Gly-Asn-Phe-Leu
> Loop's templates::-------------Asn-Trp-Phe-___-Phe-Gly-Asn
> Target:--------------Val-Ala-Asn-Trp-Tyr-___-Phe-Gly-Asn-Phe-Leu
> 
> and I want to run the following commands:
> 
> MALIGN3D
> TRANSFER XYZ ( with a CLUSTER_CUT >0)
> 
> My problems are:
> 
> 1º) As I have understood, the command MALIGN3D doesn't work because
> there aren`t equivalent positions during the framework construction
> betweeen the first template and the rest. Am I wrong?
> So I  try to solve this by superposing the loop's templates with the
> main template with the  command SUPERPOSE.

it is true, in general, but you do have an overlap, all the way (if the
above alignment is the one that is used).
Superpose is applicable to superpose two models only. For more models
you need to use MALIGN3D.

 
> 2º) After, when I run TRANFER XYZ (  CLUSTER_CUT >0), it appears in the
> .log :
> 
>             transfe_505_> MODEL is an average of the largest
> cluster:        1.0000
>             transfe_336E> No template structures.
>             recover____E> ERROR_STATUS >= STOP_ON_ERROR:        1

maybe the align codes, maybe the alignment was incorrectly defined or
not at all. 


But i would rather suggest the following. you have a general template
and two-three loop fragment templates. You want to model the whole
structure based on the template but you have a loop section in your
target, that you want to model exclusively on your fragments that you
have found, not on the general template. If I understood correctly, that
is your aim. You simply need to use an alignment like this (following
the above example, but only one loop):

Main template  ....LHGYF------VFGPTGNFL....
loopA          ---------ANWY-F-------------  
TARGET         ....LHGYFVFGPTG------NFL....
 
i.e. you misalign regions in such a way, that the target loop section is
aligned only with the loop fragment, while the rest of the target is
aligned with the rest of the template. The not used loop fragment of the
template is shifted into a gapped region, where it is not aligned with
anything. With this alignment you can use the default modeller routine
for model building.

this is only an illustration, however in practice you need to pay
attention to have a residue overlap between your loop fragment(s) and
your main template, unlike in this latter example.

Andras
> 
> I don't know what's my error.  Do I need more templates to make a
> cluster? Must the loop's templates be similar to the first template in
> order to have the same topology?
> Don't they overlap properly?
> 
> Is TRANFER XYZ (  CLUSTER_CUT =-1.0) the only possibility I have?
> (because it works)
> 
> Thanks in advance