Hi!
I have a protein sequence which is way more than 1000 aa residues. However, it does not have a crystal structure. So I obtained one from homology modeling. I used I-TASSER for this however I'm not very happy with models that have been generated. Secondly, I have to dock a small ligand to a specific residue (site- specific) and then post this, also try to dock another protein to my homology modeled protein (blind docking). However, most of the sites for protein-protein docking do not accept a protein which is more than 500 residues.
The other thing that can probably be done is, try and identify similar domains in the protein and model is separately. But then, will this will prove to be disadvantageous while predicting the right 3D structure as the folding of the entire protein will depend on the co-ordinates of every domain-like structure which was submitted separately.
I'm unsure how else to approach this. Please help.
Thanks in advance
Hi, I have similar study now, i have a sequence with 1251 aa residues. I would like to know, how to develop a reliable model using this.?
as u said , there will be an option to find the functional regions in this protein (domain) and model it separately. Can look for the domain information from the online tools, then look for the suitable template. align target sequence and template..this alignment can be taken as input for model building. for this swissmodel is one of the best software.