Skip to content

Releases: bahlolab/nf-cavalier

23.03.1

17 Mar 07:33
Compare
Choose a tag to compare

Added

  • support for mutect2 variant calls
  • pedigree validity detection and correction in cavalier package

22.07.1

05 Jul 02:12
Compare
Choose a tag to compare

Added

  • "filter stats" csv output to cavalier_wrapper.R to keep track of reasons variants are filtered

Fixed

  • Problem in cavalier::hgnc_sym2sym function causing some gene names to be incorrect

22.03.1

22 Mar 07:47
Compare
Choose a tag to compare

22.03.1 - 22 Mar 2022

Added

  • SV support, SVPV visualisation
  • Candidate variants CSV output summarising all candidate variants across all samples

Fixed

  • Test code resulting in warning 2 samples in "vcf" but not in "ped" or "bams": a_sam, b_sam removed

Changed

  • Merge cavalier-pre and cavalier containers, install the majority of R packages with conda for faster builds
  • PanelApp list versions save to cavalier cache to avoid downloading repeatedly across pipelines
  • Reorganise Nextflow code into discrete modules to simplify main.nf
  • Set Cavalier cache dir explicitly in Nextflow pipeline to avoid singularity not recognising home directory issue

21.10.2

28 Oct 23:37
0f7a59f
Compare
Choose a tag to compare

21.10.2 - 29 Oct 2021

Fixed

  • VEP cache directory not passed as Nextflow Path, causing directory not to be mounted in Singularity images.

Added

  • Warnings when sample IDs and family IDs in inputs are unmatched.
  • Splitting of input VCF based on genomic intervals to speed VCF preprocessing.
  • Param exclude_benign_missense to allow control whether benign missense variants are excluded from results.

Changed

  • Switched to BCF instead of VCF.gz format for intermediate tasks
  • Dropped unused VEP annotations --mane, --var_synonyms, --pubmed,--af_esp, --gene_phenotype, --appris,
    --tsl, --uniprot, --biotype, --canonical,--regulatory and --domains
  • Dropped VEP filter options --allow_non_variant and --dont_skip.
  • Combined vep and vep_filter processes into process vep.
  • Combined vcf_splitand vcf_flatten_multi into new process vcf_split_norm.
  • process vep now outputs 3 files: '*.vep.bcf' for annotated variants to be passed to Cavalier, '*.vep-modifier.bcf'
    for annotated variants categorised as 'MODIFIER' and '.unannotated.bcf' for variants not annotated by VEP
    (intergenic/invariant/failed reference check).