C. Titus Brown
C. Titus Brown
* gtdb * genbank * viruses * rna?
presumably mostly a problem for private genomes...
does it make any sense to compare metagenomes taxonomically (via grist output) rather than "just" via k-mers? if so, genome-grist could do that :)
Over in #130, @mr-eyes asks: >Will redundant sequence IDs affect the results? The entire header is unique but redundant IDs. and I answer: >We don't enforce anything about the IDs...
over in https://github.com/dib-lab/2020-paper-sourmash-gather/ and https://github.com/dib-lab/2021-paper-sourmash-gather-pipeline/ I've made various changes to the notebooks and figures, it'd be nice to backport the good changes to genome-grist.
@drtamermansour - >This is an example output file ``` kmers: 3096803759 known_hashes: 13349 n_bases: 72295609638 n_reads: 524035389 sample: S039 total_hashes: 3150150 unknown_hashes: 3136801 ``` >I assume “total_hashes” are the randomly...
I'm running some metatranscriptomic samples through grist, and getting very little classified with the below parameters. I'm wondering if I should lower the threshold below 5e3, and/or if I should...
right now, the base_tempdir is not ever cleaned up...
dump links here :) https://twitter.com/akorobeynikov/status/1405628344444633092
@mr-eyes asked: >And a question, does genome-grist/sourmash use the paired-end information in any step? and my answer: >Sourmash doesn’t use paired-end information in any way. >genome-grist could in the mapping...