Skip to content

Releases: Plant-Food-Research-Open/assemblyqc

Version 2.2.1

11 Dec 08:41
39eb040
Compare
Choose a tag to compare

What's Changed

Added

  1. Added notes on HTTP(s) server on the HiC page and on the need to move dynamically loaded content when moving the report's HTML file #183

Fixed

  1. Fixed an issue where PLOTSR crashed due to a mismatch in the ordering of syri.out files when synteny_plotsr_assembly_order was not specified #184
  2. Fixed an issue where a path to HiC FastQ file pairs from the current directory were considered a SRR ID #179
  3. Fixed edges and input/output arrows in the flowchart #178

Dependencies

  1. Nextflow!>=24.04.2
  2. nf-schema@2.1.1

PRs

Full Changelog: 2.2.0...2.2.1

Version 2.2.0

05 Nov 20:18
0141925
Compare
Choose a tag to compare

What's Changed

Added

  1. Added Gfastats #126
  2. Updated nf-core/template to 3.0.2 #149
  3. Updated samtools faidx to 1.21
  4. Now using nf-test for pipeline level testing #153
  5. Added text/html as content mime type for the report file #146
  6. Added a sequence labels table below the HiC contact map #147
  7. Added parameter hic_samtools_ext_args and set its default value to -F 3852 #159
  8. Added the HiC QC report to the final report so that users don't have to navigate to the results folder #162
  9. Added the fastp log to the final report #163
  10. Updated the tube map along with the tool list #166
  11. Added Orthofinder #167
  12. Changed order of tool options in the nextflow.config file
  13. Updated PFR's Kraken 2 database to k2_pluspfp_20240904 #170
  14. Increased memory requirement for Kraken 2 to 256.GB

Fixed

  1. Fixed a bug where Gene score distribution graph did not appear correctly #125
  2. Increased memory requirement for DNADIFF to avoid SLURM OOM kills with exit code 2 #141
  3. Documented the use explicit use of -revision parameter #160
  4. Now using _JAVA_OPTIONS in module RUNASSEMBLYVISUALIZER to avoid user preferences related errors

Dependencies

  1. Nextflow!>=24.04.2
  2. nf-schema@2.1.1

Deprecated

  1. Reduced the GenomeTools stats figures to 300 DPI #142
  2. Now synteny_mummer_min_bundle_size is set to 1000000 by default #142
  3. results is not the default output directory anymore
  4. Removed a number of unnecessary parameters: monochromeLogs, config_profile_contact, config_profile_url, validationFailUnrecognisedParams, validationLenientMode, validationSchemaIgnoreParams, validationShowHiddenParams validate_params
  5. Resource parameters have been removed: max_memory, max_cpus, max_time

Full Changelog: 2.1.1...2.2.0

Version 2.1.1

19 Sep 22:52
cec2728
Compare
Choose a tag to compare

What's Changed

  • Made the hic param pattern more flexible by @GallVp in #131
  • Updated modules to remove Bioconda defaults channel by @GallVp in #133
  • Fixed flowchart syntax to remove '\n' by @GallVp in #134
  • Now correctly validating gff files for circular sequences by @GallVp in #136
  • Version bump to 2.1.1 by @GallVp in #137
  • Formatted mermaid chart by @GallVp in #138
  • Fixed the branch protection GitHub action by @GallVp in #140
  • Candidate for 2.1.1 by @GallVp in #139

Full Changelog: 2.1.0...2.1.1

Version 2.1.0

02 Aug 01:33
d1c2f0a
Compare
Choose a tag to compare

What's Changed

Full Changelog: 2.0.0...2.1.0

Added

  1. Created summary presence/absence tables for NCBI FCS modules #88
  2. Added min. system requirements #91
  3. Added a test to verify the fix for the bug which resulted in a pipeline crash for assemblies without LTRs
  4. Updated NCBI FCS GX to 0.5.4 #93
  5. Updated SYRI to 1.7.0 #104
  6. Added a script to automatically check for updates on GitHub/GitLab and post issues
  7. Updated modules: UNTAR, MERYL_COUNT, GUNZIP, MINIMAP2_ALIGN, FASTQC

Fixed

  1. Fixed a bug where intron_length_distribution was used instead of cds_length_distribution when creating the CDS Length Distribution Graph #95
  2. Fixed a bug where 'Subsequent pipeline modules are skipped.' was printed in the report.html even when contamination_stops_pipeline was set to false
  3. Now NCBI FCS GX module uses all the cores available from the Nextflow task
  4. Fixed a bug which caused PLOTSR to fail for certain assembly names #102
  5. Now LTRRETRIEVER_LTRRETRIEVER does not crash when the input assembly does not contain any LTRs #92
  6. Now LTRRETRIEVER_LTRRETRIEVER does not crash when the input assembly is not writable #98
  7. Now soft masked regions are unmasked before computing LAI #117
  8. Fixed a bug in ASSEMBLATHON_STATS which caused it to fail on MMC executor due to multiple binds of the bin directory
  9. Changed NextFlow to Nextflow
  10. Updated citation to Bioinformatics

Dependencies

  1. Nextflow!>=23.04.0
  2. nf-validation@1.1.3

Deprecated

  1. Changed default branch name from master to main in nf-core template files
  2. Moved version_check.sh to .github/version_checks.sh
  3. Moved docs/contributors.sh to .github/contributors.sh
  4. Removed dependency on https://github.com/PlantandFoodResearch/nxf-modules.git
  5. Replaced nf-core/fastq_trim_fastp_fastqc with nf-core/fastq_fastqc_umitools_fastp which has nf-test unit tests
  6. Removed version check on README.md

Version 2.0.0

05 Jun 07:47
690e56b
Compare
Choose a tag to compare

What's Changed

Full Changelog: 1.4...2.0.0

Added

  1. Updated nf-core/template to 2.14.1
  2. Removed release-announcements GitHub workflow
  3. Added a list of nf-core contributors
  4. Added a launcher script for local testing local_assemblyqc
  5. Added a custom BUNDLELINKS module which respects direction when bundling DNADIFF links #82
  6. Added the ability to create linear synteny plot in addition to the circos plot #74
  7. Updated modules and sub-workflows: BWA/INDEX, BWA/MEM, CAT/CAT, , CUSTOM/RESTOREGFFIDS, CUSTOM/SHORTENFASTAIDS, GT/GFF3, GT/GFF3VALIDATOR, GT/STAT, LTRFINDER, LTRHARVEST, LTRRETRIEVER/LAI, LTRRETRIEVER/LTRRETRIEVER, SAMBLASTER, FASTA_LTRRETRIEVER_LAI, FASTQ_BWA_MEM_SAMBLASTER, GFF3_VALIDATE, CUSTOM/SRATOOLSNCBISETTINGS, FASTP, FASTQC, UNTAR, SEQKIT/SEQ, SEQKIT/SORT, FASTA_EXPLORE_SEARCH_PLOT_TIDK
  8. Now the contamination_stops_pipeline flag allows the pipeline to continue if contamination is detected. It's default value is true #54
  9. Now fasta ids are sorted in natural order for the HiC module #76
  10. Now using FASTQ_DOWNLOAD_PREFETCH_FASTERQDUMP_SRATOOLS for SRA downloads
  11. Added MERQURY module #85
  12. Replaced GFF3_VALIDATE sub-workflow with GFF3_GT_GFF3_GFF3VALIDATOR_STAT
  13. Replaced local BUSCO module with FASTA_GXF_BUSCO_PLOT sub-workflow #75
  14. Replaced local NCBI_FCS_ADAPTOR with nf-core module and updated to 0.5.0 which includes additional adaptors for PacBio and Nanopore technologies #55
  15. Added PLOTSR #77
  16. Added JADWOS01 assembly to xrefsheet for successfully running PLOTSR
  17. Now detecting duplicate sequences with SEQKIT/RMDUP #64

Fixed

  1. Fixed a bug which caused NCBI_FCS_GX to not resume #80
  2. Restored the original version of nf-core/subworkflows/fastq_trim_fastp_fastqc
  3. Fixed n-core linting
  4. Updated tower.yml
  5. Updated LICENSE copyright to Copyright (c) 2024 The New Zealand Institute for Plant and Food Research Limited #81
  6. RUNASSEMBLYVISUALIZER is now single threaded for successful execution on both Linux and MacOS
  7. Fixed java memory overflow issues in RUNASSEMBLYVISUALIZER
  8. Updated FASTA_LTRRETRIEVER_LAI to fix a pipeline crash when ch_monoploid_seqs was [ meta, [] ] #83
  9. Improved input assembly documentation #86
  10. Added assembly tag to synteny warning message regarding missing synteny_labels file
  11. Now copying files in NCBI_FCS_GX_SETUP_SAMPLE rather than symlinking in an attempt to support NextFlow Fusion

Dependencies

  1. NextFlow!>=23.04.0
  2. nf-validation@1.1.3

Deprecated

  1. Removed CIRCOS_BUNDLELINKS module
  2. Now the default value of synteny_plot_1_vs_all is false
  3. Replace module CUSTOM/CHECKGFF3FASTACORRESPONDENCE with a local groovy function in GFF3_GT_GFF3_GFF3VALIDATOR_STAT sub-workflow

Version 1.4

04 Mar 01:56
9dfb828
Compare
Choose a tag to compare

What's Changed

Added

  1. Now it is possible to skip FASTP and FASTQC for the HIC module
  2. Renamed ASSEMBLY_QC workflow to ASSEMBLYQC
  3. Now using nf-core/FASTA_EXPLORE_SEARCH_PLOT_TIDK
  4. Now redirecting validation errors to AssemblyQC report
  5. Simplified layout of CITATIONS.md file
  6. Now using pfr/gff3_validate sub-workflow for gff3 validation
  7. Now listing software versions from the versions.yml file
  8. Replaced custom GUNZIP module with nf-core/gunzip
  9. Replaced custom gt/stat with pfr/gt/stat
  10. Replaced custom fasta_validator with nf-core/fastavalidator
  11. Added pre-commit version checking
  12. Now gt/stat reports extended stats and multiple distribution plots have been added to the report
  13. Added a tools tab to the report which lists the tools used by the pipeline to create the report
  14. Refactored and cleaned data flows for all the custom sub-workflow
  15. Started using nf-core template
  16. Started using semantic versioning
  17. Moved all python depending packages to 'docker.io/gallvp/python3npkgs:v0.6'

Fixed

  1. All modules are now emitting versioning information
  2. Fixed a bug which caused LAI to run with null assembly fasta
  3. Fixed FASTA_LTRRETRIEVER_LAI sub-workflow so that it respects monoploid_ids parameter.

Dependencies

  1. NextFlow!>=23.04.0
  2. nf-validation@1.1.3

Deprecated

  1. Removed BIOCODE GFF3 STATS owing to its frequent failures