High RMSD from HADDOCK calculation

HADDOCK
1

Dear Experts,

Apology for being naive.
Please see the links to the following results from HADDOCK webserver as follows:

https://wenmr.science.uu.nl/haddock2.4/run/6113383897/75469-APP_APP_Parallel

I find here that the RMSD is hugely higher, than the suggested value (<= 10 Angs.) in one of the HADDOCK lectures (Integrative modelling of biomolecular complexes - Part 2 - YouTube).

Request you to please suggest corresponding implications of the same in my supplementary calculations.

Sincere Regards,
Prathit Chatterjee

2

The reported RMSD only tells you where a given cluster is with respect to the best model generated by HADDOCK.

It has nothing to do with the RMSD metrics used by CAPRI and in benchmarking methods (the 10A limit for ligand-RMSD).

As such it does not mean a model is good or bad

1 Like
3

Dear Bonvin,

I have several questions about docking scores:
Does RMSD value in Angstrom scale? and how should we analyze the plot of haddock score and RMSD? and what is per residue rmsd?
in the end, I have 5 protein complexes that made similar rmsd plots. what does this similarity mean?

4

I have several questions about docking scores:
Does RMSD value in Angstrom scale?

Yes

and how should we analyze the plot of haddock score and RMSD? and what is per residue rmsd?

The RMSD values are calculated with respect to the best model (best scoring model) generated.
It is just a way to visualise the data, clusters, and see if there are well-defined minima.

The RMSD itself has no predictive value. You can not compare the RMSDs of different runs.