Dear Modeller's Users:
I'm trying to model a protein dimer interacting with a small DNA molecule. Following the manual instructions I was able to build my dimer using:
>P1; my_target xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx --> first monomer / yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy --> second monomer * >P1; my_template hhhhhhhhhhhhhhhhhhhhhhhhhhhhhh --> first monomer / kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk ---> second monomer *
But I have some problems to add my DNA molecule as an additional restraint:
I added to my script:
env.io.hetatm = True
And I also added my DNA molecule (all the DNA atoms were named HETATM) into the template pdb and modified my alignment in this way:
>P1; my_target xxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxxx --> first monomer ---------- --> alignment of extra dna residues (I'm not sure is this is right) / yyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyyy --> second monomer ---------- --> alignment of extra dna residues (I'm not sure is this is right) * >P1; my_template hhhhhhhhhhhhhhhhhhhhhhhhhhhhhh --> first monomer ---------- --> alignment of extra dna residues (I'm not sure is this is right) / kkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkkk ---> second monomer ---------- --->alignment of extra dna residues (I'm not sure is this is right) / ---------------------------------------------------- ---> 1st strand of DNA ............. / ---------------------------------------------------- ---> 2nd strand of DNA ............. * Modeller worked without any warning but my dimer pass through the DNA molecule, so this is not working.
Any help would be very appreciated.
Sergio
On 1/21/11 4:11 AM, Sergio Garay wrote: > I'm trying to model a protein dimer interacting with a small DNA > molecule. Following the manual instructions I was able to build my dimer > using: ... > But I have some problems to add my DNA molecule as an additional restraint:
It's difficult to tell what your problem is without seeing a "real" alignment file. But note that you need to put the '.' residues for each DNA residue in *both* your template and your target - it looks from your post as if you only have them in the template.
Note that Modeller does know the topology of RNA and DNA, so you can use one-letter codes to get these residues in your model. However, it doesn't know anything about the secondary structure of nucleic acids, so they will probably not be modeled correctly. You are usually better off using the '.' residue, since that will force Modeller to preserve the structure from the template (and the interaction between RNA/DNA and your protein).
Ben Webb, Modeller Caretaker
-
Modeller Caretaker -
Sergio Garay