Skip to content

Releases: shenwei356/kmcp

KMCP v0.7.1

08 Feb 10:22
Compare
Choose a tag to compare

v0.7.1 - 2022-02-08 Github Releases (by Release)

  • profile:
    • new flag --metaphlan-report-version and the default value is 3. #4
    • column name renamed: from fragsFrac, fragsRelDepth, fragsRelDepthStd to chunksFrac, chunksRelDepth, chunksRelDepthStd.
    • fix computation of chunksRelDepth.
    • slightly improve sensitivity for -m 0.

KMCP V0.7.0

24 Jan 08:09
Compare
Choose a tag to compare

v0.7.0 - 2022-01-24 Github Releases (by Release)

  • commands:
    • new command utils filter: Filter search results and find species-specific queries.
    • new command utils merge-regions: Merge species/assembly-specific regions.
    • rename info to utils index-info.
  • compute:
    • skip k-mer containing Ns.
    • when splitting genome into fragments, sequences are concatenated with k-1 'N's
      instead of directly concatenation.
      It eliminates fake k-mers at the concatenation position.
    • set default value for flag -N/--ref-name-regexp: (?i)(.+)\.(f[aq](st[aq])?|fna)(.gz)?$.
    • fix a rare bug when splitting FASTQ files.
  • search:
    • support searching with paired-end reads which has a higher specificity and a lower sensitivity.
      A flag --try-se is added for search read1/read2 when the paired end reads have no hits.
    • fix matches order of a query.
    • fix queries with many Ns.
    • change default value of flag -t/--min-query-qcov from 0.6 to 0.55 (similarity ~96.5%).
    • change default value of flag -n/--keep-top-scores from 5 to 0, i.e., keep all matches by default.
    • new flag -w/--load-whole-db: load all index files into memory.
    • 10-25% faster.
    • better log.
  • merge:
    • fix adding up hits.
    • fix bug of incorrect order, reduce memory usage.
    • support one input file.
  • profile:
    • change analysis workflow, using 4 stages.
    • output format change: new column coverage, fragsRelDepth and fragsRelDepthStd.
    • change default file extension of binning file.
    • check if the taxid of a target is given by taxid mapping file.
    • automatically switch to the new taxid for a merged one.
    • change computation of score.
    • new flag -d/--max-frags-depth-stdev.
    • new option -m/--mode.
    • change default value of flag -t/--min-query-qcov from 0.6 to 0.55 (similarity ~96.5%).
    • change default value of flag -n/--keep-top-qcovs from 5 to 0 (keep all matches).
    • change default value of falg -f/--max-fpr from 0.01 to 0.05.
    • change default value of flag -H/--min-hic-ureads-qcov from 0.8 to 0.75 (similarity ~98%).
    • faster search result parsing.

kmcp v0.6.0

13 Aug 12:05
Compare
Choose a tag to compare

v0.6.0 - 2021-08-13

  • new command:
    • merge: merge search results from multiple databases.
  • compute:
    • fix splitting very short genomes.
    • remove flag -e/--exact-number, making it default.
  • index:
    • do not roundup sizes of indexes. The searching speed is not
      affected and even faster due to optimization of search command.
    • use three k-mers thresholds to control index file size.
    • better control of cocurrency number and better progress bar.
    • do not support RAMBO index anymore.
  • search:
    • 1.37X speedup, and faster for database with two or more hash functions.
    • new flag -S/--do-not-sort.
  • profile:
    • fix a nil pointer bug when no taxid mapping data given.
    • fix number of ureads.
    • new flag -m/--keep-main-matches and --max-score-gap