Skip to content

CHANGELOG.md

Latest commit

 

History

History
161 lines (122 loc) · 7.4 KB

CHANGELOG.md

File metadata and controls

161 lines (122 loc) · 7.4 KB
Raw

vsearchpipeline: Changelog

The format is based on Keep a Changelog and this project adheres to Semantic Versioning.

v1.0.0 - Mar 18 2026

Added

  • New VSEARCH_MAPPINGRATE process that calculates per-sample mapping rates from existing usearch_global outputs without re-running vsearch
  • VSEARCH_FASTQFILTER now captures stderr (*.filter_stats.txt) to record post-filter read counts per sample, used as the denominator for mapping rate
  • VSEARCH_USEARCHGLOBAL now captures stderr (mapping_stats.txt) to record overall unique-sequence mapping rate
  • Mapping rate output: mapping_rate_summary.tsv (per-sample) and mapping_rate_overall.txt (global)
  • SILVA database version now configurable via pipeline parameter (silva_db)
  • PiCrust2 stratified option added
  • Decontam module added with negative control detection and nucleic acid concentration input
  • Picrust2 module with extArgs support
  • PIPELINE_INITIALISATION and PIPELINE_COMPLETION subworkflows following modern nf-core pattern
  • Samplesheet reading and primer validation moved into PIPELINE_INITIALISATION (utils subworkflow)
  • Named VSEARCHFLOW wrapper workflow in main.nf following modern nf-core pattern
  • MultiQC skipped automatically with a warning when input exceeds 6000 FastQC files
  • GitHub Actions CI workflow (ci.yml) with matrix jobs for test and test_noprimers profiles
  • GitHub Actions benchmark workflow (benchmark.yml) triggered on release or manually; validates ASV recovery from 22-species mock community
  • GitHub Actions check_updates.yml workflow: checks Bioconda/conda-forge weekly for new tool versions and sends email notification via Gmail
  • assets/check_updates.py — Anaconda API-based version checker with bioconda → conda-forge fallback
  • assets/tools.txt — list of pipeline tool packages to monitor for updates
  • assets/versions_seen.json — tracks last known version per tool; auto-committed by the workflow
  • Stub SILVA databases for CI (assets/testdata/small/SILVA_asv_db_test.fa.gz, SILVA_species_db_test.fa.gz) filtered to taxa present in testdata
  • Stub SILVA databases for benchmark (assets/testdata/benchmark/SILVA_benchmark_asv_db.fa.gz, SILVA_benchmark_species_db.fa.gz) filtered to the 19 mock community genera
  • conf/test_noprimers.config — CI test profile without primer trimming
  • conf/test_benchmark.config — benchmark profile using mock community data with known 22-species composition
  • assets/testdata/benchmark/mock_sequences_V4.fasta — 22 reference ASV sequences for benchmark validation
  • assets/testdata/benchmark/validate_benchmark.py — validates recovered ASV count (21–23) and per-reference identity ≥ 0.97
  • docs/check_updates.md — setup instructions for the version monitoring workflow

Fixed

  • Fixed rarefaction level auto-calculation: level is now capped at max_counts so it can never exceed the deepest sample and wipe all samples; min_counts and max_counts are computed once and reused throughout
  • Fixed rarefaction user-level guard: fallback to auto-calculated level now only triggers when user_rarelevel > max_counts (i.e. no samples would survive), instead of incorrectly triggering whenever the level exceeded the shallowest sample
  • Fixed decontam bugs including column name for nucleic acid concentration (Nucl_Acid_Conc)
  • Fixed overall mapping rate not being found due to incorrect search string ("Matching unique query sequences" vs "Matching query sequences")
  • Fixed version collection for R-based processes
  • Fixed test file paths and testdata layout
  • Fixed schema for metamap input
  • Fixed small bugs in vsearch scripts
  • Fixed broken link in docs
  • Removed nf-core template boilerplate, logos and references
  • Removed unused check_max function
  • Cleaned repo structure
  • Fixed MissingMethodException crash in MultiQC caused by FastQC zip LinkedList being passed as closure arguments
  • Fixed MultiQC channel construction: zip files were incorrectly passed as --config and --logo CLI arguments
  • Removed unused ch_multiqc_config, ch_multiqc_logo, and ch_multiqc_custom_config channels
  • Removed deprecated script-level def variable declarations; all logic now inside workflow {} blocks
  • Removed NfcoreTemplate, WorkflowMain, and WorkflowVsearchpipeline Groovy helper class usage
  • Inlined INPUT_CHECK and PRIMERS_CHECK subworkflows; subworkflow files are now unused
  • Updated assets/multiqc_config.yml section IDs and links to match barbarahelena/vsearchpipeline
  • Updated assets/methods_description_template.yml with correct pipeline references and full tool citation list
  • Replaced metaboflow ASCII art in pipeline logo with VSEARCH ASCII art

Dependencies

  • Updated FastQC and MultiQC nf-core modules
  • Updated DADA2 to v1.38.0
  • Updated SILVA database to 138.2
  • Updated vsearch scripts
  • Updated mafft version

Deprecated

  • subworkflows/local/input_check.nf — replaced by PIPELINE_INITIALISATION
  • subworkflows/local/primers_check.nf — replaced by PIPELINE_INITIALISATION
  • lib/NfcoreTemplate.groovy, lib/WorkflowMain.groovy, lib/WorkflowVsearchpipeline.groovy — replaced by utils subworkflows

v0.7.1-beta - Feb 18 2024

Fixed

  • Updated FastQC and MultiQC nf-core modules to latest versions

v0.7-beta - Jan 23 2024

Added

  • New module to sort and remove singletons
  • skip_tree parameter to make phylogenetic tree construction optional
  • maxdiffpct parameter for vsearch merge pairs
  • New HPC profile with wider CPU/memory/time limits
  • error_retry label for failed jobs
  • Zenodo DOI badge

Fixed

  • Solved bugs in phyloseq modules
  • Ensured phyloseq outputs are published to outdir
  • Vertical line in rarefaction plot now placed at the actual rarefaction level
  • Fixed memory and resource labels
  • Corrected typos in documentation

Dependencies

  • Replaced IQ-TREE with FastTree for phylogenetic tree construction
  • Updated vsearch and DADA2 containers
  • Updated nf-core modules
  • Returned to SILVA 138.1 (latest stable)
  • Moved container and conda settings to modules.config

Deprecated

  • IQ-TREE option removed

v0.6-beta - Nov 6 2023

Added

  • Check for presence of primers before trimming (primer check is now optional)
  • Testdata FASTQ files without primers (*_noprimers.fastq.gz)
  • skip_primers parameter

Fixed

  • Resolved issues in metrics and rarefaction modules
  • Fixed rarefaction level fallback when all sample counts are very low
  • Restructured modules for reusability with postfix (complete / rarefied)

Dependencies

  • Updated phyloseq container

v0.5-beta - Nov 6 2023

Initial release of vsearchpipeline, created with the nf-core template.

Added

  • Full VSEARCH-based amplicon pipeline: merge pairs, quality filter, dereplication, chimera removal, clustering (UNOISE3), taxonomy assignment (DADA2 / SILVA), phyloseq object construction
  • Optional phylogenetic tree (FastTree / IQ-TREE) and MSA (MAFFT)
  • Optional PiCrust2 functional prediction
  • Rarefaction and diversity/composition metrics modules
  • Decontam module for negative-control-based contaminant removal
  • MultiQC and FastQC quality control
  • nf-core-compatible samplesheet and primer sheet input
  • Test profile with small testdata
  • HPC (Snellius) profile

Dependencies

  • VSEARCH
  • DADA2
  • phyloseq
  • SILVA 138.1
  • FastTree / MAFFT
  • PiCrust2
  • FastQC / MultiQC