Hi
I used HADDOCK webserver to generate models based on unambiguous restraints. For publishing the results, sampling precision and cluster precision are required. Are these parameters already included in the HADDOCK result downloaded? I don’t know which score HADDOCK provided might represent the precision of models.
Thank you very much.
Not sure what you are meaning by sampling and cluster precision…
I would for sure report the sampling done for docking, i.e. the number of it0/it1/water models generated.
Also report how many clusters have been generated. And ideally report in a table the stats of the various clusters, e.g.
Those are all provided in the results page.
Important, if you select a given cluster, it should be significantly better than the other, or you should have some additional data to based your choice on.
Thank you for the advice. We will report the stats provided on the result page. As for the number of models generated at each step, I did not locate it on the result page. Where should I find this information?
The sampling precision and model/cluster precision are defined as below(https://pubmed.ncbi.nlm.nih.gov/31150619/):
Precision or uncertainty of a model or ensemble of models: a measure of variability of the ensemble of models
Sampling precision: granularity of sampling used to find models consistent with input information
From what I understand, the sampling precision would a value to represent the differences of all the models and the model/cluster precision would be the differences within the cluster I picked. I wonder whether there’s any similar values provided by HADDOCK as alternatives. Thank you.
The default setting of the server can be found at:
https://bianca.science.uu.nl/haddock2.4/settings
In you did not change the sampling, the number of models generated for the various stages is:
- it0 (rigid body): sampling 10’000, 1’000 written to disk
- it1 and water refinement: 200