I have 3 X-ray structures with co-crystallized ligand in active site. How should I actively copy all ligand in this same position from all temple to the final homology models? I perform homology modelling in MODELLER 9v12. When I transfer one ligand to my protein everything works fine, but when I try transfer all ligand, I get unfolded model with all ligand in random position in active site. Should I change something in script?
Hello
One way is to build a model with the template with co-crystallized ligand. for this purpose you have to change the alignment file so that it can recognize the ligand.
Put dot (.) in the alignment file and then build a model. Here dot represents the heteroatom apart from amino acid residues. use this alignment file and build the model.
check the tutorial http://salilab.org/modeller/tutorial/advanced.html
If you don't want to change the alignment file then after building homology model align the model with the template pdb in pymol and copy the coordinates of ligands and save as a pdb file.
i hope it helps you.
best wishes,
Gaurao
Hello
One way is to build a model with the template with co-crystallized ligand. for this purpose you have to change the alignment file so that it can recognize the ligand.
Put dot (.) in the alignment file and then build a model. Here dot represents the heteroatom apart from amino acid residues. use this alignment file and build the model.
check the tutorial http://salilab.org/modeller/tutorial/advanced.html
If you don't want to change the alignment file then after building homology model align the model with the template pdb in pymol and copy the coordinates of ligands and save as a pdb file.
i hope it helps you.
best wishes,
Gaurao
Gaurao's reply is just on the spot.
adding (.) in the alignment for each hetatm would be the prefered and easiest option. Just remember to allow your script to work with hetatms with:
env.io.hetatm = True
and that should work.
The second option he mentions is also OK, but it would require extra steps and you would loose any optimisation of the side-chain that MODELLER might do.
Cheers,
Jaume
If proteins are already overimposed or structurally well aligned, there is many option. Here are two. In graphic software is usually possible to select the ligand and protein chains you like independently and then to save them in a separate file.
A second option is to cut the coordinates of both structures (ligand and homology model) from the original ascii files, by copying and paste them into a new file. This new ascii file should be readeable by the graphic software.
- Badges
- Science topic
- Similar topics
- Bioinformatics
- Bioinformatics and Computational Biology