Output possible losses in assignment (id2taxa mapping or Bio4j lookup)

ohnosequences / mg7

Configurable and scalable 16S metagenomics data analysis

https://goo.gl/y3rZFD

GNU Affero General Public License v3.0

3 stars 3 forks source link

Output possible losses in assignment (id2taxa mapping or Bio4j lookup) #59

Closed laughedelic closed 8 years ago

laughedelic commented 8 years ago

Strange result in the test:

all assignments have to be under Bacteria(2) or Archaea(2157)
but the results of direct counts are
- 100% root
- 100% Cellular organisms (direct ancestor of Bacteria and Archaea)
- ~50% Bacteria

laughedelic commented 8 years ago

I've tested this and the added lost-* outputs are always empty (as it should be). So the problem is somewhere else.

laughedelic commented 8 years ago

I'm merging this. This is not necessary, because if the database is built correctly, these lost-* outputs will be always free. But for the time while we're testing MG7, we can see that if they appear, something is wrong.