Hmm, I think it would make more sense to have here some nested for loops instead of iterating multiple times through the list.

I think on something like this (untested):

users = []
for analysis in self.Analyses:
    for verificators in analysis.getVerificators():
        for userid in verificators:
            if not userid:
                continue
            user = api.get_user(userid)
            if not user:
                continue
            if user in users:
                continue
            users.append(user)

What do you think?

IMO three nested fors are uglier than this "flatten" approach. Note that you are iterating through the list of users inside the loop each time you get a user, while I simply remove duplicates of userids before looping.

I think that since we don't expect these lists of verifiers to be huge (in most cases, 1 per sample and most samples will share same verifier), the performance impact caused by iterating three times is negligible. I can remove the filter(None, userids) and perform the check inside the loop though.