Hello
I have a protein and RNA structure and when I wanted to run it I noticed Optimised run for bioinformatics predictions because I wanted the best structure I left this option on during docking and to have a comparison I ran the same structure in the off mode. In the on mode, HADDOCKscor showed a lower value and when I checked the two structures I noticed that they were docked differently I wanted to know which mode is more suitable to use (I am planning to mutate the structure) (I used another online docking server to find the binding region selection and used the results obtained in HADDOCK I selected the same binding region in both structures)
Hi,
the option " changes some parameters of the execution machinery (full list is here Settings for running with bioinformatic predictions – Bonvin Lab). Basically, HADDOCK samples and refines more models to compensate for the supposedly low quality of the restraints (bioinformatic prediction restraints are usually of lower quality than experimental restraints).
In this case, everything depends on your restraints: if you, as you said, used another server to infer the binding region, then I’d say that turning this parameter on makes sense. Of course, if the defined region is wrong, the output models will be wrong as well
Cheers,
Marco
And to add to Marco’s answer: Never consider only the top ranked cluster. Look at the other solutions.
Are their score significantly different from the top one?
Thanks
What do you suggest for choosing the right binding region?
Thanks
In opt mode -58.3 +/- 4.2 and in other mode -72.4 +/- 1.4, the binding region and orientation also differed.
In opt mode, the top clusters differ by 2,3 units and in other mode, the top clusters are very different
as far as I know for protein-RNA this is still pretty much an open problem…you can try to
- dig into the literature about your protein of interest
- use AlphaFold3
- check out the interfaces formed by your protein (for example using ARCTIC-3D Arctic3D Webserver).
Blind protein-RNA interaction prediction is a real challenge!
Thank you for your guidance and time.