My selected target has an antibody as its ligand. However, my compounds of interest are not proteins/peptides so I'm having difficulty in solving this problem because I cannot use two different softwares to compare the binding affinity of the compounds and the antibody towards the target. can someone help me thanks!
Haddock ( https://wenmr.science.uu.nl/haddock2.4/ )
"HADDOCK distinguishes itself from ab-initio docking methods in the fact that it encodes information from identified or predicted protein interfaces in ambiguous interaction restraints (AIRs) to drive the docking process. It also allows to define specific unambiguous distance restraints (e.g. from MS cross-links) and supports a variety of other experimental data including NMR residual dipolar couplings, pseudo contact shifts and cryo-EM maps. HADDOCK can deal with a large class of modeling problems including protein-protein, protein-nucleic acids and protein-ligand complexes, including multi-bodies (N>2) assemblies."
Hi Yuhta, I have no prior wet lab experiment. I'm trying to do in silico studies of my compounds with respect to ERBB2 but the ligand attached to it is also a protein Transtuzumab Yuhta Nomura
Annemarie Honegger Anuj Kumar Thanks for your help! Do you have guidelines for the optimum size of grid box? thanks!!
I suggest you have a look at some of the tutorials:
https://www.bonvinlab.org/software/bpg/proteins/#tutorials
thanks Annemarie Honegger Also I would like to ask if it is okay to use a mutant protein in molecular docking? I mean in protein data bank there are entries that indicate that there are mutations. Are there any preparations/modifications that I need to do prior to docking if that is the case? I mean aside from removing water, repairing missing side chains, adding kollman charges. thanks!
Depends on what question you want to answer with your simulation, and where th mutations in your protein are. E.g. most GPCR structures in the pdb contain stabilising mutations, since the wild-type proteins are only produced at very low levels, and are not stable enough to survive crystallisation. They may affect conformational equilibria, and either push the receptor to the active or to the inactive conformation, pre-disposing it to agonist or to antagonist binding.
Normally, the publication associated with the structure should indicate whether or not the mutations affect ligand binding. If the mutations are close to the putative ligand binding site, and you are worried that they affect ligand binding directly, you can use the mutant structure to model the wild-type sequence.
thank you Annemarie Honegger this enlightened me! by creating the model does this mean homology modeling?
Depending on the number and location of substitutions, you can either just replace the individual residues with a mutagenesis tool, e.g. the Mutagenesis Wizard in PyMol (as long as you do not have any insertions/deletions), or use a homology modelling site or program, e.g. Modeller under UCSF Chimera
hello Annemarie Honegger thanks this is of great help! I opened the Mutagenesis wizard in pymol and I now know how to change a particular residue to the desired one using the app. However, I tried to look at the protein in rcsb.org and the mutations are at the lower numbered residue. For example 2OZ7, the androgen receptor has a mutation at position 207, however, the protein structure starts at sequence # 671, hence I cannot change position 207. How to go about this? Thank you for your response!
For the human androgen receptor (https://www.uniprot.org/uniprot/P10275), only the C-terminal ligand binding domain has been crystallised (residues 670-920).
Much of the N-terminal sequence is intrinsically disordered, or, in the case of the DNA binding domain, may only assume a defined structure upon interaction with its DNA target sequence. Residue 207 is within a region that is predicted to be intrinsically disordered. ( https://www.uniprot.org/uniprot/P10275 )
If you are looking for compounds competing with the natural ligand, you can do your docking to the isolated ligand binding domain.
However, I would first do a comparison of the different available structures for this domain, to assess to what degree the different ligands in these structures affect the shape of the binding pocket, and, depending on the results, use several different ones as docking targets if there are clear differences.
hi Annemarie Honegger thanks for the input!! This makes sense! With regard to the comparison of the different structures, does that mean I obtain all the crystallized structures of the receptor that are available and compare the docking results?
You do a structural superposition of all templates and look at the structural differences and check functional annotations (associated publications, looking up the ligands in corresponding databases.
For receptors, which by definition are signalling molecules, you usually see conformational differences depending on whether the ligand binding domain is binding to an agonist (ligand which activates the receptor) or to an antagonist (ligand which stabilises the inactive conformation and prevents the natural ligand from binding and activating the receptor. For a well-studied example of the conformational changes involved in signalling by nuclear receptors, look at the estrogen receptor (e.g PDB molecule of the Month # 45 "Estrogen Receptor", Article Estrogen Receptor
).Which conformational state is the right on to use in your case depends on the questions you wish to answer with your docking experiment. Here are some review articles to familiarise yourself with what is known about the structure and function of the androgen receptor:
https://www.nature.com/articles/aps201418
Article Androgen Receptor Structure, Function and Biology: From Benc...
Article Structural Insights of Transcriptionally Active, Full-Length...
If the mechanism of inhibition is the same as in the Estrogen receptor, you may not be able to dock an antagonist into the receptor conformation from an agonist-bound structure due to sterical hindrance.
For this problem you can try Schrödinger software, you can find your answer as well.
- Badges
- Science topic
- Similar topics
- Biological Science
- Immunology
- Antibodies