On May 22, 2009, at 11:10 PM, Ben Webb wrote:
> Keren Lasker wrote: >> It is just not fully clear to me if em_fit should be part of helper >> or em ? > > Well, it was a hypothetical application, but... neither.
it is not a hypothetical one - because I am actually trying to wrap it up these days .... ;)
> Calculation of > an EM fit should go in the IMP.em module, of course. But if you wanted > to write a little command line application that took a PDB and a > density > map from the user and reported the CCF, for example, that would go in > the 'applications' directory, not in the IMP.em module. (The important > point is that the bulk of the functionality - i.e. reading a PDB file, > reading an EM map, calculating the CCF - should stay in IMP.atom and > IMP.em where it currently is, so that other applications or modules > can > use it.) I don't think anything EM-related should go in IMP.helper, > since we don't want to end up with too many interdependencies between > the modules. yes - I just was not sure where we draw the line between stuff that should be helper vs. in modules. I think that Daniel meant for the modules to have just a very basic functionality, although I agree that rigid fitting should go in EM. I guess that the decision would be on a case-to-case basis, for example flexible fitting should probably go directly to application and not to em. > > > Ben > -- > ben@salilab.org http://salilab.org/~ben/ > "It is a capital mistake to theorize before one has data." > - Sir Arthur Conan Doyle > _______________________________________________ > IMP-dev mailing list > IMP-dev@salilab.org > https://salilab.org/mailman/listinfo/imp-dev