4Mashima

=ON the distribution of Feature vs qualifier relation = Measure the recurrence distribution of qualifiers under each feature. Recurrence distribution:
 * Under F1: q1x1 = 100 records, q1x2=1000records,...,q1x1025=1 records (max)
 * F1:q2 (once=1640000 records, twice=1 record (max))
 * F1:q3 (once=0, twice=0, 5times=100, ..24times=1 (max))

ex) How many times qualifier organism is used under one source feature?

average recurrence, range =1-100, SD=xxx　may be OK as answer.