ohnosequences / db.rna16s

A comprehensive, compact, and automatically curated 16S database

GNU Affero General Public License v3.0

8 stars 0 forks source link

[MOVE TO db.rna16s.assignments] Assign a set of taxonomy nodes to each reference (and precompute their LCA) #85

Closed agarciamontoro closed 6 years ago

agarciamontoro commented 6 years ago

Each reference in the DB needs to be assigned to a set of nodes in the taxonomy tree from ohnosequences/ncbitaxonomy repo. We may also compute the LCA of this set.

We need to:

[ ] Document how to assign a sequence in the DB to a node in the taxonomy. Where is this information stored? How do we parse it?
[ ] Document what to do in the cases a reference is assigned to each of the subtrees specified in ohnosequences/ncbitaxonomy#7.
[ ] Code a function Reference -> Seq[TaxonomyNode] that implements the assignment.
[ ] Decide how to store this information in this DB.
[ ] Evaluate the function over all references and store the results with the precomputed LCA in the DB.

@eparejatobes, can you take charge of the documentation tasks?

Needs ohnosequences/ncbitaxonomy#7.

eparejatobes commented 6 years ago

Sure I can do this. This should be moved to a different repo/project though, otherwise we are forced to update the db itself everytime we change/correct assignments.

eparejatobes commented 6 years ago

that was the point of #72 (no description there, my fault)

agarciamontoro commented 6 years ago

This should be moved to a different repo/project

Are you ok if I move this to assignments.rna16s?

that was the point of #72 (no description there, my fault)

72 was closed in #76 by error, then? Or did I miss something there that solved this?

eparejatobes commented 6 years ago

72 is solved inasmuch as assignments are no longer here; we still need to put them elsewhere.

Are you ok if I move this to assignments.rna16s?

as these assignments will be checked/fixed (clustering, taxonomy tree coherence) I'd rather have a separate repo for them.

agarciamontoro commented 6 years ago

I'd rather have a separate repo for them.

Sure! I'll create a repo later called db.rna16s.assignments (someday we should review all names and find a consistent way of generating them :p) and move this issue there.

I'll close this when that repo is ready.

eparejatobes commented 6 years ago

someday we should review all names and find a consistent way of generating them

They basically correspond to packages

agarciamontoro commented 6 years ago

Closed in favour of https://github.com/ohnosequences/db.rna16s.assignments/issues/2.