Skip to content

Releases: nf-core/taxprofiler

v1.2.4 - Bouncy Basenji Patch

28 Aug 14:07
1385e2f
Compare
Choose a tag to compare

Added

  • #619 Added default pipeline nf-test (added by @sofstam)
  • #615 Updated to nf-core pipeline template v3.3.1 (added by @sofstam)
  • #614 Updated module krakenuniq/preloadedkrakenuniq to fix preload mechanism. The RAM chunk size can now be customised on a per-database basis. (added by @muniheart with help from / ❤️ to @jfy133)
  • #600 Added metaphlan to multiqc report (❤️ to @harper357 for reporting and added by @sofstam)
  • #605 Added documentation on how to generate Bracken reports for different taxonomic levels (❤️ to @MajoroMask for reporting and added by @jfy133)

Fixed

  • #602 Fixed a typo in the run-merging table example in usage.md (❤️ to @ilight1542 for reporting and fixed by @jfy133)
  • #592 improved JSON schema for better validation (by @microlei)
  • #604 KrakenUniq no longer fails due to pathname splitting during batching when only single sample supplied (by @AlexHoratio)
  • #612 Fix bug with bracken results being shown in MuktiQC for long read data every if bracken not supported for long reads (added by @jfy133 and @sofstam )
  • #630 Fix a bug in KRAKENUNIQ reads channel where a single-element list containing a / in the path was being split incorrectly, causing unexpected multiple list elements (❤️ to @sysbiocoder for reporting and fixed by @LilyAnderssonLee)
  • #631 For paired-end reads, DIAMOND will use only the forward read file (read 1) unless reads are merged (❤️ to @sysbiocoder for reporting and fixed by @LilyAnderssonLee)
  • #635 Fix faulty parsing of memory request in BBDuk (❤️ to @Proton-NTA for reporting, fix by @mahesh-panchal)

Changed

  • #602 Rename (undocumented) test_nothing profile to test_minimal for consistency with other pipelines (by @jfy133)
  • #614 There is now no default krakenuniq_ram_chunk_size value for KrakenUniq. By default now the entire database will be loaded into memory unless you use the parameter, or specify --preload-size in a database sheet. (added by @muniheart with help from / ❤️ to @jfy133)
  • #624 Updated all relevant nf-core modules to latest version (by @sofstam)

Dependencies

Tool Previous version New version
Nextflow 24.10.0 25.04.2
bbduk 39.06 39.18
bowtie2 2.5.2 2.5.4
bracken 2.9 3.1
diamond 2.1.8 2.1.12
fastp 0.23.0 0.24.0
ganon 2.0.0 2.1.0
kraken2 2.1.3 2.1.5
minimap2 2.28 2.29
motus 3.0.3 3.1.0
multiqc 1.27 1.30

v1.2.3 - Bouncy Basenji Patch

13 Mar 08:42
970bdc2
Compare
Choose a tag to compare

Added

  • #576 Sort input to krakenuniq to enable retrieval of cached batch tasks (❤️ to @muniheart for reporting and fix)
  • #569 Updated to nf-core pipeline template v3.1.2 (added by @sofstam)
  • #571 Updated to nf-core pipeline template v3.2.0 (added by @sofstam)

Fixed

  • #580 Minor documentation typo fixes (❤️ to @AlexHoratio for reporting and fix)
  • #573 Improved help messages and documentation to state many of the taxpasta related params require taxonomy files to be input (❤️ to @alexhbnr for reporting, fix by @jfy133)
  • #578 Fixed incorrect keyword in nextflow_schema.json (fix by @mashehu)
  • #585 Fixed code blocks (fix by @masheshu)
  • #586 Fixed wrong argument name for saving analysis reads (fix by @epstein6)

v1.2.2 - Bouncy Basenji Patch

15 Jan 09:38
80d2fcc
Compare
Choose a tag to compare

Added

Fixed

  • #561 Updated Centrifuge to fix internal error when running with conda (❤️ to @AnotherSimon for reporting, fix by @jfy133)

v1.2.1 - Bouncy Basenji Patch [2024-11-28]

28 Nov 15:09
0af416a
Compare
Choose a tag to compare

Added

Fixed

  • #550 Fixed a bug where single-end data with the same sample names that include . in it would overwrite output files (❤️ to @SannaAb for reporting, fixed by @Midnighter)
  • #552 Fixed the issue where nonpareil fails due to insufficient data for calculation caused by a missing PNG (❤️ to @Joon-Klaps for reporting, fixed by @LilyAnderssonLee)

Dependencies

Deprecated

v1.2.0 - Bouncy Basenji [2024-10-03]

03 Oct 11:43
6893894
Compare
Choose a tag to compare

Added

Fixed

  • #518 Fixed a bug where Oxford Nanopore FASTA input files would not be processed (❤️ to @ikarls for reporting, fixed by @jfy133)
  • #523 Removed hardcoded -m lca from GANON_CLASSIFY due to more options in new version of ganon (fixed by @LilyAnderssonLee & @jfy133)
  • #531 Fix FASTA input validation in schema allowing FASTQ extension, expand allowed FASTA extensions (fixed by @jfy133)
  • #512 Minor formatting and ordering improvements in MultiQC report (by @jfy133)
  • #532 - Added missing documentation behind the 'ignore' BRACKEN_BRACKEN error strategy (❤️ to @Mavti for reporting, fixed by @jfy133)
  • #536 - Redefine contents_re for filtlong to fix its missing from the MultiQC report (fixed by @LilyAnderssonLee)

Dependencies

Tool Previous version New version
bbmap 39.01 39.06
bowtie2 2.4.4 2.5.2
bracken 2.7 2.9
diamond 2.0.15 2.1.8
ganon 1.5.1 2.0.0
kraken2 2.1.2 2.1.3
krona 2.8 2.8.1
megan 6.24.20 6.25.9
metaphlan 4.0.6 4.1.1
minimap2 2.24 2.28
motus 3.0.3 3.1.0
multiqc 1.21 1.25
samtools 1.17 1.20

Deprecated

v1.1.8 - Augmented Akita Patch [2024-06-20]

20 Jun 15:59
5d3ee55
Compare
Choose a tag to compare

Added

  • #487 Updated to nf-core pipeline template v2.14.1 (added by @jfy133)

Fixed

  • #484 Improved input validation to immediately fail if run accession IDs within a given sample ID are not unique (❤️ to @sofstam for reporting, fixed by @jfy133)
  • #491 Added flag to publish intermediate bracken files (❤️ to @ewissel for reporting, fixed by @sofstam and @jfy133)
  • #489 Fix KrakenUniq classified reads output format mismatch (❤️ to @SannaAb for reporting, fixed by @jfy133)
  • #495 Stop TAXPASTA failures when profiles do not have exact compositionality (fixes by @Midnighter, @jfy133)

Dependencies

Tool Previous version New version
KMCP 0.9.1 0.9.4
TAXPASTA 0.6.1 0.7.0

Deprecated

  • #492 Removed --kmcp_mode parameter from KMCP to allow per database specification by setting in db_params in database sheet (fixed by @jfy133)

v1.1.7 - Augmented Akita Patch [2024-04-29]

29 Apr 08:08
f1e6153
Compare
Choose a tag to compare

Added

  • #477 Provide more emphasis and links to tutorials on how to retrieve and supply reference databases (❤️ to @vmkalbskopf for reporting, added by @jfy133)

Fixed

  • #476 Fixed bug in validating Bracken/Kraken/KMCP split database parameters (fixed by @LilyAnderssonLee)

Dependencies

Deprecated

v1.1.6 - Augmented Akita Patch [2024-04-16]

16 Apr 10:54
45ce748
Compare
Choose a tag to compare

Added

  • #454 Updated to nf-core pipeline template v2.13.1 (added by @LilyAnderssonLee & @sofstam)
  • #461 Turned on 'strict' Nextflow evaluation runs (added by @jfy133)
  • #461 Optimised database compression so each compressed input database is untarred once, and shared amongst each run with different parameters (added by @jfy133)
  • #461 Added new parameter to optionally save uncompressed databases (added by @jfy133)
  • #471 Removed -stub run in the download_pipeline.yml because the pipeline does not support stub runs on dev (fixed by @LilyAnderssonLee)

Fixed

  • #336 Replaced samplesheet check with nf-validation for both sample and database input sheets (fix by @LilyAnderssonLee)
  • #460 Corrected the channel transformations to combine Kaiju and mOTUs reports with their reference databases (fix by @Midnighter)

v1.1.5 - Augmented Akita Patch [2024-02-08]

08 Feb 10:13
9f47401
Compare
Choose a tag to compare

Added

  • #439 Read deduplication with fastp (added by @maxibor)
  • #440 Include mention of pre-built kaiju databases in tutorial.md (added by @Joon-Klaps)
  • #442 Updated to nf-core pipeline template v2.12 (added by @sofstam)

Fixed

  • #444 Centrifuge now uses dedicated tmp directory to hopefully prevent mkfifo clashes (❤️ to @erinyoung for reporting, fix by @jfy133)

Dependencies

Tool Previous version New version
Centrifuge 1.0.4_beta 1.0.4.1

v1.1.4 - Augmented Akita Patch [2024-01-24]

25 Jan 09:08
21e6af8
Compare
Choose a tag to compare

Fixed

Dependencies

Tool Previous version New version
kaiju 1.8.2 1.10.0