Hello,
I have a sequence and a structure aligned in a file in RIP format. Now I want to just copy the coordinates of our structure to our sequence without any changes. I am using transfer-xyz.py but it doesn't copy the coordinates as they are in the structure before or after the loops.
Is there a way to transfer the coordinates from structure to sequence as they are?
Regards,
Ali Algarrous
Ali Algarrous wrote: > I have a sequence and a structure aligned in a file in RIP format. > Now I want to just copy the coordinates of our structure to our > sequence without any changes. I am using transfer-xyz.py but it > doesn't copy the coordinates as they are in the structure before or > after the loops. > > Is there a way to transfer the coordinates from structure to sequence > as they are?
transfer_xyz will copy coordinates of equivalent atoms in aligned residues, so if you want your model to have exactly the same coordinates as your template structure, you will need a 1:1 (gapless) alignment. If you have gaps/loops in your alignment, obviously Modeller will not be able to copy the coordinates since none exist.
What is it that you're trying to do?
Ben Webb, Modeller Caretaker
Thank you so much for your response. This is also what I was expecting that transfer_xyz is not going to do what I want.
Let's say I have something like this:
>P1;1a1t structureX:1a1t:FIRST:A:LAST:A:::: MQKGNFRNQRKTVKCFNCGK---EGHIAKNCRAPRKKGCWKCGKEGHQMKDCTERQAN---*
>P1;madeup sequence:madeup:::::::: MGKGNFRMQRKTVKCFNCGKD--EGHIAKNCRMNCDKGCWKCGKEGHQMKDCTERQANGRT*
I want Modeller to copy the coordinates of the template to my sequence for at least (N CA C CB) atoms of each residue. And if there is a gap/loop, then Modeller should generate the coordinates for it.
Is there anyway that Modeller would do what I want?
Thank you in advance.
Regards, Ali Algarrous
-----Original Message----- >From Modeller Caretaker modeller-care@salilab.org Sent Sat 6/30/2007 9:10 PM To Ali Algarrous algarra@uni.edu Cc modeller_usage@salilab.org Subject Re: [modeller_usage] Help with transfer_xyz method
Ali Algarrous wrote: > I have a sequence and a structure aligned in a file in RIP format. > Now I want to just copy the coordinates of our structure to our > sequence without any changes. I am using transfer-xyz.py but it > doesn't copy the coordinates as they are in the structure before or > after the loops. > > Is there a way to transfer the coordinates from structure to sequence > as they are?
transfer_xyz will copy coordinates of equivalent atoms in aligned residues, so if you want your model to have exactly the same coordinates as your template structure, you will need a 1:1 (gapless) alignment. If you have gaps/loops in your alignment, obviously Modeller will not be able to copy the coordinates since none exist.
What is it that you're trying to do?
Ben Webb, Modeller Caretaker
Ali Algarrous wrote: > Thank you so much for your response. This is also what I was > expecting that transfer_xyz is not going to do what I want. > > Let's say I have something like this: >> P1;1a1t > structureX:1a1t:FIRST:A:LAST:A:::: > MQKGNFRNQRKTVKCFNCGK---EGHIAKNCRAPRKKGCWKCGKEGHQMKDCTERQAN---* > >> P1;madeup > sequence:madeup:::::::: > MGKGNFRMQRKTVKCFNCGKD--EGHIAKNCRMNCDKGCWKCGKEGHQMKDCTERQANGRT* >
> I want Modeller to copy the coordinates of the template to my > sequence for at least (N CA C CB) atoms of each residue.
Sure - that's exactly what model.transfer_xyz() will do.
> And if there > is a gap/loop, then Modeller should generate the coordinates for it. > > Is there anyway that Modeller would do what I want?
Absolutely, but that's a separate step, handled by a separate command - model.build(). If you look at the example script (examples/commands/transfer_xyz.py) you'll see that model.build() is called immediately after model.transfer_xyz() to do this. This will also build the coordinates for sidechains if the residue differs between target and template.
Ben Webb, Modeller Caretaker
-
Ali Algarrous -
Modeller Caretaker