To evaluate herbicidal activity, I'm planning to perform molecular docking between plant Acetyl-CoA Carboxylase (ACCase) and various natural compounds. I studied that ACCase consists of Biotin Carboxylase, Biotin Carboxyl Carrier Protein, and Carboxyltransferase, and herbicides predominantly bind to the Carboxyltransferase domain.
For molecular docking of this enzyme, what are the key differences between using a full-length homology model versus a Carboxyltransferase-domain–only homology model? Which is better solution?
If there is little difference, is it acceptable to perform docking using only the specific domain that contains the binding site?
I would appreciate it if you could answer my questions.
Dear Hyeonseon Kang,
The answer to your question is closely related to the location of the binding site you are considering. You mentioned that some compounds bind to the Carboxyltransferase domain. Do you have more specific information about the binding site location—namely, where they bind within the Carboxyltransferase domain?
If the compounds bind at the interface between this domain and another, it would be inappropriate to use only the Carboxyltransferase domain. On the other hand, if the binding site is distant from other domains, the Carboxyltransferase domain alone is probably sufficient for your docking simulation.
If you don’t yet have this information, I suggest thoroughly reviewing the literature for experimental data (e.g., mutagenesis or structural studies) that might help you identify the binding site. If no such data are available and your goal is to perform a global docking to identify the binding location, I strongly recommend using a full-length structure.
Naomi Scarano
Thank you for your answer. :)
The ACCase I intend to model is from Arabidopsis thaliana. While its precise binding site remains uncharacterized, UniProt provides the sequence for the carboxyltransferase domain within the full ACCase sequence.
Additionally, I used the AlphaFold-predicted 3D structure to select a template from PDB. After performing structural alignment between the AlphaFold structure and the PDB template (specifically choosing templates with pesticide-bound ligands), I confirmed that the ligand-binding regions overlayed accurately.
Based on this, I propose designating this overlapping site as the binding pocket.
Given this context, I am uncertain whether to proceed with global homology modeling or local homology modeling.
- Badges
- Science topic
- Similar topics
- Methodology
- Research Methodology
- Research Papers