v0.8.1

Full Changelog: v0.8.0...v0.8.1

Two minor bug fixes:

• Ignore tree leaves absent from the index: if a given placement tree (via -t) has leaves that are not among the indexed reference genomes, distances and placements remain unaffected.
• Fixed a bug that resulted in double semicolons in the jplace placement tree output.

v0.8.0

• Changed the behavior of --dist-max. It no longer triggers --filter. The distances can be filtered using --dist-max without limiting the output to distances that are indistinguishable from the minimum distance.

• Fixed the corrupted jplaces due to a stream synchronization issue

v0.7.3

• Fixed minor input parameter validation bug
• Fixed corrupted place due to a stream synchronization issue

v0.7.2

Full Changelog: v0.7.1...v0.7.2
Mostly bug fixes:

• Unclosed jplace when there is no placement.
• Add a color when the nonce is nonzero.
• Set node hash after the parsing finishes.
• Dangling pointers due to reallocation when the tree is highly unresolved.
• Make sure to output NaN when no match.

v0.7.1

• Added Dockerfile
• Minor bug fix for cases when -t is not used
• Important bug fix for LSH position drop for reading old indexes
• Improved inspect command for displaying the details for indexes
• Added HLL based and more accurate rho estimation
• Slightly better sdust handling for subsampling rate, still not very accurate

Full Changelog: v0.6.0...v0.7.1

v0.6.0

This is also a Bioconda release.

• Tabular reporting option.
• Parsing GTDB/Greengenes style lineages for taxonomic placement.
• Better robust newick parsing and a stable jplace tree.
• Added --summarize option to aggregate all placements or distance-based hits.
• Changed defaults (including multiplacement).

Full Changelog: v0.5.2...v0.6.0

v0.5.2

Full Changelog: v0.5.1...v0.5.2

v0.5.1

For Bioconda -- Full Changelog: v0.5.0...v0.5.1

v0.5.0

New features:

• Chi-sq option to control placemet algorithm, and faster placement algorithm.
• Saving index metadata in text format for inspection.
• Specify a tree other than the one used during indexing.
• Added multiplacement and best distance reporting.
• Added pext for fast bit extraction when bmi2 is available.

Bug fixes:

• Incorrect pext masking resulted in wrong HD.
• Choosing the closest match histogram with respect to the reverse complement.
• Newick parsing with quotation marks.
• Handling of Ns in queries by ignoring.

Full Changelog: v0.4.3...v0.5.0

v0.4.3

• Important bug fixes thanks to @armandiB (issue #6).
  i) Having an input reference sequence shorter than the k-mer (or the minimizer) length triggered a bug which resulted in some of the reference sequences being ignored.
  ii) Setting -w and -k to the same value led to unexpected behavior.

• Significant speed up due to better parallelism, especially with higher number of threads.