Topological correctness #54
Conversation
fixtures/aa.gfa
Outdated
| S 1 A SN:Z:123 SO:i:0 SR:i:0 | ||
| S 2 A SN:Z:123 SO:i:0 SR:i:0 | ||
| L 1 + 2 + * | ||
| P 124 1+,2+ 4M |
There was a problem hiding this comment.
i'm not sure what the parser would do, but this should be 1M (1 match)
There was a problem hiding this comment.
Why? Shouldn't it be 0M for overlap of length 0 if it's representing the sequence AA?
There was a problem hiding this comment.
yeah, it should be. I knew 4m was way too much :)
src/imports/fasta.rs
Outdated
| @@ -57,13 +64,16 @@ pub fn import_fasta( | |||
| .sequence(&sequence) | |||
| .save(conn) | |||
| }; | |||
| let node = Node::create(conn, &seq.hash); | |||
There was a problem hiding this comment.
So if i import the same file 2x, we'll have duplicate nodes + edges made?
There was a problem hiding this comment.
Yeah. It's not great, but I don't see it as blocking. I think properly detecting the duplicates in general would take some work, worth doing long term.
There was a problem hiding this comment.
Is it that much of a problem on the fasta import side? When would people import the same fasta file and expect them to collapse onto the same nodes and edges? I can understand on the vcf import you'd eventually want a heuristic that says something like "for a pair of edges (p, q), if there is already a pair of edges (a, b) with a shared node id (from the reference path) and identical coordinates on all side, then check to see if the new node has the same hash as the old node."
src/models/node.rs
Outdated
| ); | ||
| let _ = conn.execute(&insert_statement, ()); | ||
| Node { | ||
| id: conn.last_insert_rowid() as i32, |
There was a problem hiding this comment.
is there a reason to do this over returning id?
There was a problem hiding this comment.
I genuinely thought I was using the full node somewhere, but I guess not. Changed
dkhofer
force-pushed
the
topological-correctness
branch
from
a930c4e
to
283dad8
Compare
No description provided.