[MRG] add column 3 to kreport

[bluegenes]

Fixes #2305

Updates kreport to include information in column 3, which is the bp "assigned" to a particular taxon. Since we make all assignments to the genome level and then apply taxonomy later, this column just contains the species-level assignments from column 2. I think this best mimics the intention of the output format / outputs from other tools.

Updated documentation with this and the recommendation to generate sketches with abundance.

This is basically a bug fix, since this is closer to what the format should have been to begin with.

[bluegenes]

I think this best mimics the intention of the output format / outputs from other tools.

My only thought here is that while sourmash assigns k-mers to genomes, the actual level of our resolution depends on alphabet and k-mer size. We don't do taxonomic assignments any differently, but I wouldn't necessarily recommend using, e.g. protein k7, to get species-level assignments. Perhaps I should add a sentence about this in the doc? e.g.

Note: sourmash makes all assignments to genomes, then integrates taxonomy information and
uses LCA-style summarization to build assignments. For species-level specificity, our current
recommendation is to use use our default k-mer size of 31.

I'm not really sure though, because 1. this is really just general sourmash info/recommendation, and 2. there are lots of reasons to try other ksizes.

[codecov]

Codecov Report

Merging #2306 (66416a6) into latest (e615ce0) will increase coverage by 7.30%.
The diff coverage is 100.00%.

@@            Coverage Diff             @@
##           latest    #2306      +/-   ##
==========================================
+ Coverage   84.84%   92.14%   +7.30%     
==========================================
  Files         131      100      -31     
  Lines       15687    11416    -4271     
  Branches     2189     2190       +1     
==========================================
- Hits        13309    10519    -2790     
+ Misses       2083      602    -1481     
  Partials      295      295              

Flag	Coverage Δ
python	92.14% <100.00%> (+<0.01%)	⬆️
rust	?

Flags with carried forward coverage won't be shown. Click here to find out more.

Impacted Files	Coverage Δ
src/sourmash/tax/tax_utils.py	98.32% <100.00%> (+<0.01%)	⬆️
src/core/src/index/search.rs
src/core/src/signature.rs
src/core/src/ffi/index/revindex.rs
src/core/src/lib.rs
src/core/src/ffi/storage.rs
src/core/tests/test.rs
src/core/src/index/mod.rs
src/core/src/index/sbt/mod.rs
src/core/src/index/linear.rs
... and 22 more

📣 We’re building smart automated test selection to slash your CI/CD build times. Learn more

[bluegenes]

ok @ctb @sourmash-bio/devs ready for review

[ctb]

I think this best mimics the intention of the output format / outputs from other tools.

My only thought here is that while sourmash assigns k-mers to genomes, the actual level of our resolution depends on alphabet and k-mer size. We don't do taxonomic assignments any differently, but I wouldn't necessarily recommend using, e.g. protein k7, to get species-level assignments. Perhaps I should add a sentence about this in the doc? e.g.

Note: sourmash makes all assignments to genomes, then integrates taxonomy information and
uses LCA-style summarization to build assignments. For species-level specificity, our current
recommendation is to use use our default k-mer size of 31.

I'm not really sure though, because 1. this is really just general sourmash info/recommendation, and 2. there are lots of reasons to try other ksizes.

I like the idea of adding this text, actually. We're seeing more and more evidence that people are reading, sometimes scouring, the text for details - let's give it to 'em!

The one addition I would make is to say "sourmash gather makes all assignments to genomes, and then sourmash tax integrates taxonomy information...".