Hello MODELLERs,
I am new to MODELLER and would like to know if there is an easy way to accomplish this task. I have a raw sequence that I want to align with my 3D structural alignment that contains nine sequences. First, it seems like my raw sequence should be formatted so that MODELLER can read it. Since this is just a sequence with no corresponding structure shouldn't the second line of the sequence file just be:
sequence:
It seems strange to me that in the manual all the examples of sequences have a corresponding PDB code when the point of homology modelling is to create the model hence no PDB code. Or for this do I just make one up?
If the sequence file for the target is then correct can I do the following:
READ_ALIGNMENT FILE = 'structures3d.ali', ALIGN_CODES = 'all' # my 3D # structural alignment for nine structures
SET ADD_SEQUENCE = on, ALIGN_BLOCK = NUMB_OF_SEQUENCES
READ_ALIGNMENT FILE = 'sequence.ali', ALIGN_CODES = 'all' # my target # sequence
SET OVERHANG = 0 ALIGN2D GAP_PENALTIES_1D= -400 -150, GAP_PENALTIES_2D = 0.5 0.5 0.5 0.5 0.5 6
Any advice would be much appreciated. General strategies that accomplish the same basic task but don't address my specific problem are also welcome.
Troy Wymore Biomedical Initiatives Group Pittsburgh Supercomputing Center
Troy Wymore wrote: > > Hello MODELLERs, > > I am new to MODELLER and would like to know if there is an easy way to > accomplish this task. I have a raw sequence that I want to align with my > 3D structural alignment that contains nine sequences. > First, it seems like my raw sequence should be formatted so that MODELLER > can read it. Since this is just a sequence with no corresponding structure > shouldn't the second line of the sequence file just be: > > sequence: > > It seems strange to me that in the manual all the examples of sequences > have a corresponding PDB code when the point of homology modelling is to > create the model hence no PDB code. Or for this do I just make one up?
Sure, you can.
> > If the sequence file for the target is then correct can I do the > following:
So far I do not see any objections to that, as long as the gaps in seq...ali are make sense in the context of the str...ali
The only thing I don't understand is that why would you need to do that, when both structures and sequence can happily live in one alignment for the MODELER purposes?
(Advantage of keeping them in one file is to be able to see easily the quality of str-seq alignment in any editor with fixed size fonts)...
My two cents,
Azat.
> > READ_ALIGNMENT FILE = 'structures3d.ali', ALIGN_CODES = 'all' # my 3D > # structural alignment for nine structures > > SET ADD_SEQUENCE = on, ALIGN_BLOCK = NUMB_OF_SEQUENCES > > READ_ALIGNMENT FILE = 'sequence.ali', ALIGN_CODES = 'all' # my target > # sequence > > SET OVERHANG = 0 > ALIGN2D GAP_PENALTIES_1D= -400 -150, GAP_PENALTIES_2D = 0.5 0.5 0.5 0.5 > 0.5 6 >
-- - Azat Badretdinov, Ph.D. - Molecular Simulations Inc. - 9685 Scranton Road - San Diego CA 92121-3752 - Phone: (858) 799 5728
Dear Troy,
> First, it seems like my raw sequence should be formatted so that MODELLER > can read it. Since this is just a sequence with no corresponding structure > shouldn't the second line of the sequence file just be: > > sequence: >
Here is the sample sequence from the manual : >P1;1fdx sequence:1fdx: 1 : : 54 : :ferredoxin:Peptococcus aerogenes: 2.00:-1.00 AYVINDSC--IACGACKPECPVNIIQGS--IYAIDADSCIDCGSCASVCPVGAPNPED*
For a sequence that you call 1dum of say 98 aa line1 should be: >P1;dum1 What you need in line 2 is just structureM:dum1: 1: : 98: :dummy: extra terrestrial:0.0:0.0 <--optional parameters> (structureM for model) You can thus read in a complete sequence but build a model for only part of the entire length.
The format is enhances PIR. There are format converters that can write out a pir format sequence. You can use ClustalX for this.
You then need to edit the protein name in line 1 and put in the information in line 2.
Hope this helps,
Shoba ======================================================================= Shoba Ranganathan, Ph D Tel:+61-2-9351-1870 ANGIS, Bay 16, Suite 104 Toll free:1-800-728-028 Australian Technology Park Fax:+61-2-9351-1878 Eveleigh NSW 1430 Australia Personal URL: http://molmod.angis.org.au/~shoba ================MolMod Home Page: http://molmod.angis.org.au===========
participants (3)
-
Azat Badretdinov -
Shoba Ranganathan -
Troy Wymore