-
Notifications
You must be signed in to change notification settings - Fork 79
New issue
Have a question about this project? Sign up for a free GitHub account to open an issue and contact its maintainers and the community.
By clicking “Sign up for GitHub”, you agree to our terms of service and privacy statement. We’ll occasionally send you account related emails.
Already on GitHub? Sign in to your account
sbt_search error: mismatch in max hash; comparison fail #146
Comments
hmm, I thought I'd calculated the k-31 signatures using --with-cardinality.
Can you extract the leaf node that is causing the problem and look at its
YAML to see if the cardinality is in there? Check with @luizirber on the
simplest way to extract the YAML filename.
|
bump. |
@ctb I think this is a bug in sbt_search unless sbt_search and sbt_gather use scaled signatures in a different manner. I'm using the same sbt with sbt_search and sbt_gather. When I calculate a signature with --scaled 10000 and use sbt_search I get an error but sbt_gather returns the best match as expected. In contrast, if I don't calculate the signature with scaled sbt_search returns several matches. Same behavior with the -k 21, 31, and 51 sbt's. I will poke @luizirber about this. |
sounds reasonable. can you build a small test case (i.e. using only a few .sig files and putting them in an SBT) that generates the error? I bet if you calculate the signatures with |
@ctb worked just like that. Also, gather doesn't work when the query signatures are calculated with cardinality. Notebook https://github.com/brooksph/2017_sbt_search_project/blob/master/dev/Untitled.ipynb |
I'm getting a similar error. How can I fix this?
|
Hi Josh,
thank you for such a complete error report!!
The '--seed' argument for compute needs to match the seed used to
construct the signatures in the database - that's 42, per mash default.
There's no reason to set the seed to anything in normal operations,
note.
best,
--titus
|
When using sbt_search with microbes.k31.sbt.json sourmash spits out an error. I calculated a signature for a single genome using sourmash compute -k 31 --scaled 10000 *fna.gz. The output of this works fine with sbt_gather but fails with sbt_search.
../dev/sourmash sbt_search -k 31 --dna microbes.k31.sbt.json ../fasta/herpetosiphon_geysericola.fa.gz.sig --threshold=0.001 --best-only
The text was updated successfully, but these errors were encountered: