nf-core/ampliseq version 2.6.0 - 2023-06-27

Summary of changes

• Updated taxonomic databases (UNITE 9.0, PR2 5.0.0)
• Added more choice for taxonomic classification: phylogenetic placement, SINTAX
• Codon usage based filtering allows for additional QC of coding regions
• Add NF-TEST pipeline end-to-end tests for existing CI tests
• nf-core template update from 2.7.2 to 2.8

Detailed changes

Added

• #564,#567,#582 - Added phylogenetic placement
• #577 - Added SINTAX for taxonomic classification
• #575, #586 - Added filtering step for stop codons for ASVs that are of coding regions.
• #580 - Add NF-TEST pipeline end-to-end tests for existing CI tests
• #591 - New version of the Unite taxonomy databases: 9.0
• #596 - New version of the PR2 taxonomy database: 5.0.0, only available with DADA2 (--dada_ref_taxonomy)
• #597 - Samples with less reads than specified with --min_read_counts (default: 1) stop the pipeline, previously the threshold was 1KB in size.

Changed

• #563 - Renamed DADA2 taxonomic classification files to include the chosen reference taxonomy abbreviation.
• #567 - Renamed --dada_tax_agglom_min and --qiime_tax_agglom_min to --tax_agglom_min and --dada_tax_agglom_max and --qiime_tax_agglom_max to --tax_agglom_max
• #580 - GitHub Actions CI - pull_request to dev tests with NXF_VER latest-everything & pull_request to master tests with NXF_VER 22.10.1 & latest-everything
• #598 - Updated Workflow figure with SINTAX and phylogenetic placement
• #599 - For exact species assignment (DADA2’s addSpecies) PR2 taxonomy database (e.g. --dada_ref_taxonomy pr2) now excludes any taxa that end with ” sp.”.

Fixed

• #553 - Handle empty barrnap results files
• #554 - Accept taxonomy strings that contain #,'
• #569 - Make header of overall_summary.tsv consistent between input data types
• #573 - Avoid parser error for single-end data when an empty read file is detected
• #578 - Template update for nf-core/tools version 2.8, including changing System.exit(1) to Nextflow.error()
• #594 - Update metadata documentation
• #595 - Closing gaps in rarefaction depth for diversity calculations (mindepth in QIIME2_DIVERSITY_CORE)

nf-core/ampliseq version 2.5.0 - 2023-03-02

Summary of changes

• Added COIDB DADA2 taxonomic database
• Small adjustments to taxonomic classification, ASV filtering by barrnap, fasta input
• sbdi-export works for more taxonomic databases now
• nf-core template update from 2.6 to 2.7.2

Detailed changes

Added

• #518,#534 - Add COIDB DADA2 reference taxonomy database
• #521 - Export svg in addition to pdf files for quality plots from DADA2
• #538 - Parameter --diversity_rarefaction_depth controls the minimum rarefaction depth for diversity analysis, this allows increasing the rarefaction depth at the cost of excluding low count samples. Parameter --ancom_sample_min_count sets the minimum sample counts to retain a sample for ANCOM analysis.

Changed

• #537 - Update output generated with option sbdi-export
• #541 - Remove adjustments of taxonomic levels for RDP & SILVA & GTDB & UNITE database for DADA2 taxonomic classification, reduced default of --dada_tax_agglom_max from 7 to 6
• #548 - --filter_ssu accepted any barrnap hit to a kingdom (domain) (any occurence in resulting gff) to choose an ASV, now only ASVs with the kingdom (domain) that has the lowest evalue are accepted.

Fixed

• #513 - Template update for nf-core/tools version 2.7.2
• #519 - Adding the pipeline reference to the MultiQC report
• #520,#530 - Fix conda packages
• #531,#546 - Update documentation
• #535 - Make sure barrnap runs with fasta input
• #544 - Adding module to fix header in fasta input if needed

Dependencies

• #528 - Updated QIIME2

Removed

• #513 - Removed parameter --enable_conda.

nf-core/ampliseq version 2.4.1 - 2022-12-07

Summary of changes

Most importantly this release fixes a bug in 2.4.0 that prevented the use of the SILVA v138 database with QIIME2.

Detailed changes

Added

• #494 - --metadata_category_barplot accepts a comma separated list of metadata categories and plots for each barplots with average relative abundance.

Changed

• #492 - --qiime_adonis_formula accepts a comma separated list of formulas.

Fixed

• #486 - Fixed typo in error message stating --skip_classifer instead of --classifier.
• #487,#488 - Update stale links in usage documentation.
• #489 - Reduce linting warnings for nf-core tools version 2.5.1.
• #491 - Make output from —addSH match UNITE format by replacing spaces with underscores.
• #495 - Template update for nf-core/tools version 2.6
• #501 - Check for empty fields in samplesheet column “run” and raise an appropriate error.
• #503 - Changed environment for formatting databases.
• #504 - Fixed warnings with nextflow 22.10 (and later) about processes that are defined more than once.

Dependencies

nf-core/ampliseq version 2.4.0 - 2022-09-07

Summary of changes

• Custom DADA2 reference taxonomy databases are now supported.
• Illumina NovaSeq data can be correctly processed.
• ASV length filter is available for stricter QC.
• Species classification is now reported in columns “Species” using DADA2’s assignTaxonomy function or “Species_exact” using DADA2’s addSpecies function, the latter only assigns exact sequence matches.

Detailed changes

Added

• #456 - An optional ASV length filter can be activated using --min_len_asv <int> and/or --max_len_asv <int>.
• #458 - Samplesheet, ASV fasta file, and/or metadata sheet is now exported into <results>/input/
• #459 - MIDORI2 CO1 database with keys midori2-co1=gb250 and midori2-co1 for --dada_ref_taxonomy
• #460 - Taxonomic ranks for DADA2 taxonomic classification can be now adjusted using --dada_assign_taxlevels <comma separated string>.
• #461 - A custom DADA2 reference taxonomy database can now be used with --dada_ref_tax_custom and --dada_ref_tax_custom_sp, typically accompanied by --dada_assign_taxlevels.
• #446,#467 - Binned quality scores from Illumina NovaSeq data can be now corrected with --illumina_novaseq.
• #477 - QC plots of DADA2’s plotQualityProfile are now also produced after preprocessing.
• #478 - Added GTDB R07-RS207 DADA2 taxonomy reference databases

Changed

• #444,#457,#463,#465,#466,#469 - Updated the documentation.
• #445 - The minimum number of total bases to use for error rate learning by default is 1e8 (DADA2, learnErrors, nbases). Previously, samples were read in the provided order until enough reads were obtained (DADA2, learnErrors, randomize=FALSE). Now, samples are picked at random from those provided (DADA2, learnError, randomize=TRUE) and a seed is set.
• #453 - Export a few more basic QIIME2 fragments (zipped files) that can be easily imported into the correct QIIME2 version for custom analysis.
• #464 - Reported taxonomic classifications on species level based on DADA2’s assignTaxonomy (approximations) is now listed in column “Species” while exact matches based on DADA2’s addSpecies are now reported in column “Species_exact”.

Fixed

• #448 - Updated SBDI export scripts to include Unite species hypothesis information if available.
• #451 - Pairwise statistics will be now performed on a subset of metadata columns specified with --metadata_category instead of ignoring that setting.
• #451 - Replace busybox with Ubuntu base image for GCP support.
• #455 - Stop with descriptive error when only one of --trunclenf and --trunclenr is given, earlier it was silently ignored.
• #474 - Template update for nf-core/tools version 2.5.1
• #475 - Report software versions for DADA2_TAXONOMY

Dependencies

• #479 - Updated software

nf-core/ampliseq version 2.3.2 - 2022-05-27

Summary of changes

This release fixes bugs and adds more options to manipulate cutadapt parameters.

Detailed changes

Added

• #429 - --cutadapt_min_overlap sets cutadapt’s global minimum overlap (-O) and --cutadapt_max_error_rate sets cutadapt’s global maximum error rate (-e) for trimming primer sequences.
• #431 - --skip_dada_quality allows to skip quality check with DADA2. This is only allowed when --trunclenf and --trunclenr are set.

Changed

• #432 - The number of records to sample from a fastq file was decreased from 5e+06 to 5e+04 for plotQualityProfile (DADA2_QUALITY), therefore a smaller subset of reads is sampled for determining --trunlenf and --trunclenr. This should make the process more robust also from larger data sets.

Fixed

• #428 - Fixed samplesheet sampleID entries, now allows dashes.
• #433 - Fixed typos and improved documentation layout.
• #437 - Template update for nf-core/tools version 2.4
• #439 - Fixed a bug in DADA2_QUALITY process with large number of nucleotides.

nf-core/ampliseq version 2.3.1 - 2022-04-05

Summary of changes

This release fixes a bug.

Detailed changes

Fixed

• #415 - ADONIS test was running by default in version 2.2.0 and 2.3.0 on metadata columns that did not always have the required format, now specifying --qiime_adonis_formula is required to run that step.

nf-core/ampliseq version 2.3.0 - 2022-04-04

Summary of changes

• Added Barrnap that annotates whether ASV sequences are SSUs and filter on demand
• Primer sequences are now not required any more and cutadapt can be skipped
• Updated to nf-core template 2.3.2
• Fixed some bugs

Detailed changes

Added

• #385 - --skip_cutadapt allows to skip primer trimmimg.
• #390, #408 - Add values to option --cut_its, specifying which ITS part to use. Also, add option --its_partial <x> to allow partial ITS sequences longer than given cutoff.
• #395 - --seed specifies the random seed.
• #396 - Barrnap annotates ASV sequences for SSU’s, it can be skipped with --skip_barrnap. --filter_ssu takes a comma separated list of “bac,arc,mito,euk” and enables SSU filtering depending on Barrnap (default: off).
• #397 - Complement README.md with links to the nf-core bytesize 25 (nf-core/ampliseq).

Changed

• #385 - --FW_primer and --RV_primer are not obligatory any more, however primer sequences are still required with cutadapt (i.e. without --skip_cutadapt), --qiime_ref_taxonomy, and --cut_dada_ref_taxonomy (cuts reference with primer sequences).

Fixed

• #384 - For QIIME2 beta diversity, make directory before execution.
• #394 - Prevent simultaneous usage of --qiime_ref_taxonomy and --classifier.
• #402, #410 - Template update for nf-core/tools version 2.3.2
• #403 - Limit number of files for DADA2_QUALITY (plotQualityProfile) by read numbers.
• #405 - Fix omitting taxonomic filtering (QIIME2_FILTERTAXA).
• #404 - Rephrase error message for empty input files and empty files after trimming with cutadapt.
• #407 - Fix reformatting script for Unite.

Dependencies

nf-core/ampliseq version 2.2.0 - 2022-01-31

Summary of changes

• Added features (e.g. SBDI-GTDB db update, Adonis test)
• Added convenience updates (e.g. properly reporting and on demand ignoring low read samples)
• Updated to nf-core template 2.2
• Fixed some bugs

Detailed changes

Added

• #352, #372 - --skip_dada_addspecies allows to skip species level classification to reduce memory requirements, incompatible with --sbdiexport that expect species annotation.
• #364 - Adonis in QIIME2 for testing feature importance in beta diversity distances, --qiime_adonis_formula can be set to provide a custom formula.
• #366 - New version of the SBDI-GTDB taxonomy database: v. 3. (Fixes problem with Reverse_ added to some domain strings.)

Changed

• #354 - Input files and files after primer trimming with cutadapt are required to be >1KB (i.e. not empty) and either the pipeline will stop if at least one sample file fails or the failing samples will be ignored when using --ignore_empty_input_files or --ignore_failed_trimming, respectively.
• #376 - Forbid sampleIDs starting with a number when also --metadata is used, because such strings are unintentionally modified and the metadata will not match any more.

Fixed

• #377- An error message will occur when --sbdiexport is used with --skip_taxonomy or --skip_dada_addspecies
• #375- Updated documentation regarding not using curly brackets in --extension with --single_end
• #362- Template update for nf-core/tools version 2.2, now requires nextflow version >= 21.10.3
• #374- Cutadapt results can be now also viwed in the MultiQC report

Dependencies

Removed

• #350 - Remove redundant derepFastq step (has no impact on results)

nf-core/ampliseq version 2.1.1 - 2021-10-28

Summary of changes

This release fixes bugs.

Detailed changes

Added

• #336 - Taxa agglomeration levels with --dada_tax_agglom_min, --dada_tax_agglom_max, --qiime_tax_agglom_min, --qiime_tax_agglom_max, with defaults that go to genus level for abundance tables and ANCOM analysis

Changed

• 338 - Write empty space instead of NA for missing values in output files.
• 342 - Added PICRUSt2 to summary figure.

Fixed

• #329 - Improve error message when no data files are found
• #330 - Make --skip_fastqc usable again
• #339 - Fix sample names when using --double_primer or --illumina_pe_its

nf-core/ampliseq version 2.1.0 “Gray Steel Boa” - 2021-09-14

Summary of changes

• Updated documentation
• Added export to the Swedish Biodiversity Infrastructure (SBDI) on demand (--sbdiexport)
• Added newer versions of reference databases
• Predict community functions on demand (--picrust) for 16S rRNA amplicons with PICRUSt2
• Fixed some bugs

Detailed changes

Added

• #322 - Export to the Swedish Biodiversity Infrastructure (SBDI)
• #294 - New version of the PR2 taxonomy database: 4.14.0, contains also some Bacteria, see PR2 release 4.14.0 notes
• #294 - New version of the Unite taxonomy databases: 8.3
• #302 - Pipeline workflow figure in README.md
• #307 - Functional predictions with PICRUSt2, on demand with --picrust
• #310 - Workflow figure in usage.md when using --multiple_sequencing_runs
• #312 - Added curated GTDB 16S taxonomy: sbdi-gtdb as parameter to --dada_ref_taxonomy
• #318 - Output information about the used reference taxonomy in a separate file in results folder dada2/ or qiime2/taxonomy

Changed

• #313 - Relative abundance tables in qiime2/rel_abundance_tables/ on ASV level were renamed and with either DADA2 (rel-table-ASV_with-DADA2-tax.tsv) or QIIME2 classifications (rel-table-ASV_with-QIIME2-tax.tsv), if available.

Fixed

• #306 - Sample names can now be identical to basenames of read files
• #299, #301- Template update for nf-core/tools version 2.1
• #303 - Reverse primer of PacBio and IonTorrent reads should be now given to --RV_primer in usual direction (before: reverse complement)
• #305 - --max_len now accepts integers as expected
• #314, #315 - ASV fasta input via —input fixed (was broken in 2.0.0) and a test profile was added.

Dependencies

• #299 - Updated MultiQC to v1.10
• #319 - Updated DADA2 from 1.18.0 to 1.20.0

nf-core/ampliseq version 2.0.0 “Blue Copper Kangaroo” - 2021-06-29

Summary of changes

• Re-wrote whole pipeline in nextflow DSL2 instead of DSL1
• Three types of input is now supported: data folder, sample sheet, ASV fasta file
• ASV generation is now performed by DADA2 in R instead of DADA2 in QIIME2
• More reference taxonomy databases are now supported out of the box (SILVA, UNITE, PR2, GTDB, RDP)
• More data types can be analysed (Illumina SE & PE, PacBio, IonTorrent)

Detailed changes

Added

• #229 - --single_end for single-ended Illumina data
• #229, #245, #267 - Taxonomic classification with DADA2
• #229 - --dada_ref_taxonomy for taxonomic classification with DADA2’s assignTaxonomy and addSpecies functions
• #278 - --qiime_ref_taxonomy for taxonomic classification with QIIME2
• #239 - Support of RDP database for DADA2 classification
• #237 - Support of UNITE database for DADA2 classification
• #229 - --input may point (1) at a fasta file ending with .fasta/.fna/.fa that will be taxonomically classified, (2) at a samples sheet ending with .tsv that allows analysis of multiple sequencing runs by reading the optional column run, or (3) at a folder input
• #229 - --sample_inference, --concatenate_reads, --illumina_pe_its; please check the documentation for their function
• #275 - Read count summary
• #274 - --skip_qiime to prevent any steps that are executed with QIIME2
• #272 - --cut_its to cut ASV sequence to ITS region before performing taxonomic classification with DADA2
• #280 - Added support for IonTorrent data
• #283 - --cut_dada_ref_taxonomy allows extracting expected amplicons from DADA2 reference taxonomy database

Changed

• #254 - Updated CamelCase parameters to be lower_case_snake_case:
  • multipleSequencingRuns to multiple_sequencing_runs
  • minLen to min_len
  • maxLen to max_len
  • maxEE to max_ee
• #277 - Requires nextflow version >= 21.04.0

Fixed

• #273 - Template update for nf-core/tools version 1.14

Dependencies

• #272 - New dependency ITSx v1.1.3
• #229 - Updated from cutadapt v2.8 to v3.2
• #229 - Updated DADA2 from v1.10 to v1.18.0, now not using QIIME2 for ASV generation any more
• #229 - Updated QIIME2 to v2021.2

Removed

• #229 - --manifest is superseeded by --input that can now also handle a sample sheet file input (required extension: .tsv)
• #229 - --Q2imported and untilQ2import are removed because pausing at that point is not neccessary
• #229 - --split is no longer supported, therefore all sample IDs have to be unique
• #229 - --classifier_removeHash and --qiime_timezone became unnecessary
• #229 - --onlyDenoising is deprecated in favour of --skip_taxonomy (which does the exact same thing)
• --taxon_reference became unnecessary
• #229 - --reference_database and --dereplication are not supported any more. --qiime_ref_taxonomy allows now choosing a taxonomic reference

nf-core/ampliseq version 1.2.0 “Teal Bronze Lion” - 2021

Added

• #106 - Added support for PacBio data
• Added --taxon_reference to be able to support both ‘silva’ and ‘unite’
• #157 - Added possibility to run double cutadapt steps, --double_primer
• #211 - Added quality filter option --maxEE

Fixed

• #182 - Fix input in case there are no underscores in sample IDs
• #186 - Update github actions
• #187 - Sample ids are in incorrect order in feature-table from PacBio data
• #201 - Template update for nf-core/tools version 1.12.1
• #147 - Split make_classifier in two different processes that can be allocated different resources
• #183 - Don’t fetch taxonomy/create classifier when run with --skip_taxonomy
• #180 - MultiQC, cutadapt and fastQC now work with --multipleSequencingRuns

Dependencies

• Updated from cutadapt v2.6 to v2.8

nf-core/ampliseq version 1.1.3 - 2020

Added

• #170 - Cite paper for initial release
• #111 - Added parameter for user specified manifest file
• #118 - Added social preview images
• #135 - Added --trunc_rmin to make sure that auto trunc cutoff retaines a certain fraction of reads

Fixed

• #172 - Template update for nf-core/tools v1.11
• #163 - Template update for nf-core/tools v1.10.2
• #136 - Pipeline fails with remote working directory
• #152 - Don’t fetch taxonomy/create classifier when run with --onlyDenoising

Dependencies

• Updated from MultiQC v1.6 to v1.9

Deprecated

• --reads is replaced by --input due to nf-core/tools v1.10.2

nf-core/ampliseq version 1.1.2 - 2019

• No further changes, except a bugfix for the timezone issue found by @marchoeppner
• Specification of ‘—qiime_timezone’ might be required to run the analysis appropriately

nf-core/ampliseq version 1.1.1 - 2019

Pipeline Updates

• Update from QIIME2 v2018.6 to v2019.10, including DADA2 v1.6 to DADA2 v1.10

Bugfixes

• #78 - All sequenced classified to the same species

nf-core/ampliseq version 1.1.0 “Silver Lime Bee” - 2019

Pipeline updates

• #40 - Added support for data originating from multiple sequencing runs
• #53 - DADA2 report is always exported
• #49 - Allowed more filtering options
• #5 - Introduced check for existence of input files
• Extended parameter sanity check, including #15
• #61 - Improved documentation
• #62 - Utilize nf-core/configs centrally for this pipeline
• #63 - QIIME imports files by using a manifest, giving more freedom with input file names
• #84 - Add proper nf-core logo

Bug fixes

• #57 - Indicate exact regex for sequencing file names
• #60 - publish demux.qza when —untilQ2import

Initial release of nfcore/ampliseq!

The pipeline is a bioinformatics analysis pipeline used for 16S rRNA amplicon sequencing data.

The workflow processes raw data from FastQ inputs (FastQC), trims primer sequences from the reads (Cutadapt), imports data into QIIME2, generates amplicon sequencing variants (ASV, DADA2), classifies features against the SILVA v132 database, excludes unwanted taxa, produces absolute and relative feature/taxa count tables and plots, plots alpha rarefaction curves, computes alpha and beta diversity indices and plots thereof, and finally calls differentially abundant taxa (ANCOM). See the output documentation for more details of the results.