Use fit_model for large-scale screening

[DrrDom]

Hello, I see this method to compare two molecules, can we use this to do large-scale screening?
(I don't know what the load_from_feature_coords means and why the output is a tuple)

Pharmacophore match
Create a two-point pharmacophore model and match with a pharmacophore of a molecule (both pharmacophores should have identical binning steps)

q = P()
q.load_from_feature_coords([('a', (3.17, -0.23, 0.24)), ('D', (-2.51, -1.28, -1.14))])
p.fit_model(q)
Output

(0, 1)

Originally posted by @RJ-Li in #5 (comment)

[DrrDom]

@RJ-Li, could you please provide a reproducible piece of code? Then I'll be able to explain what it does.

The answer on your general question is yes. This function fit_model is used for matching of a query and a model. The output, I expected, should be a dictionary, not a plain tuple. Nevertheless, these are ids of features of a model matching a query. However I have to aware that it will be not that fast as the best tools. You also may look at this implementation - https://github.com/meddwl/psearch/blob/gen_pharms/psearch/screen_db.py