Releases: broadinstitute/viral-pipelines
Releases · broadinstitute/viral-pipelines
v2.1.8.0
Update to Dockstore 1.9 API
- direct github apps based integration for all branch commits and tagged releases to viral-pipelines to Dockstore (skips viral-ngs-staging intermediary) using their latest API released a few days ago [#139, #140]
Workflow updates:
- fixes to NCBI Fetch_SRA_to_BAM [#138]
- new workflow classify_single (a single-sample version of classify_multi) [#133]
- new mafft_and_snp [#132, #133]
- replace all
augur align
calls with our own mafft invocation that is much faster [#132] - augur_from_msa can take multiple keep_list inputs [#134]
- allow kraken2 task to take fasta input, not just bam (to allow for running on contigs) [#127]
Misc:
v2.1.4.1
v2.1.0.4
assemble_refbased workflow:
- default to minimap2 aligner instead of novoalign [#123, #110]
- add run_discordance task and output metrics for measuring discordance between sequencing runs within a sample [#124]
- disable ivar trim quality trimming due to unidirectional trim behavior (only run ivar trim for PCR primers) [#122]
- remove MultiQC from refbased workflow [#82]
metagenomics:
- bugfixes for kraken2 build [#120, #117]
- bugfixes for kaiju task [#113]
- remove blastx (optional) tasks from classify_multi [#102, #106]
BEAST (GPU):
- increase dnanexus runtime limit and specify dx_instance type [#116, #118]
- update docker image [#111]
nextstrain / phylogenetics:
- add
snp-sites
task to emit vcf files [#115] - add
augur mask
step [#99] - VM shape tuning [#112]
- rename and reorganize workflows a bit so they are all called
augur_from_xxx
[#121, #101]
miscellaneous:
v2.1.0.2
v2.1.0.1
v2.1.0.0
- assemble_refbased changes [#83, #84]:
- bugfix: workflows should never fail when no genome is produced. Empty output files are now properly emitted with no errors.
- increase default min_coverage for consensus calling from 2 major allele reads to 3
- bump ivar docker images to 1.2.2
- update documentation
- build_augur_tree changes [#85, #90]
- add
augur clades
task and integrate with main workflow - integrate existing
augur filter
task with main workflow - increase
AUGUR_RECURSION_LIMIT
to handle very deep/large trees - update documentation
- add
- filter_bam_to_taxa bugfix for
taxonomic_ids
optional input [#92] - add
align_and_count_multiple_report
for improved/bugfixed spike in count handling [#91] - VM runtime reshaping [#88, #89, #92]
- disable GCP preemption for certain long running tasks
- expose disk size requirements for certain tasks and increase their default values
- increase default RAM for
rmdup_ubam
task
- update viral-core (and downstream docker images) to v2.1.0
v2.0.21.5
- added vcf merging workflow and tasks (with snpEff annotation) [#61, #77]
- update dnanexus default kraken2 database [#80]
- Travis: bugfix for dockstore & viral-ngs-staging for tagged releases failing to sync [#78]
- Travis: reduce the number of CI tests for cromwell-on-local-travis instead of running the full set (for build speedups)
- update to viral-phylo 2.0.21.5
- genbank workflow updates [#79, #81]:
- switch from mafft MSA-based alignment before annotation transfer to a scattered pairwise alignment prior to annotation transfer[#76]
- add travis test coverage for both single-segment and multi-segment genomes to cover both code paths
- RTD documentation updates
v2.0.21.4
- new workflows for beast: beast_gpu, beast_to_auspice [#66, #74]
- new workflows: subsample_by_metadata and newick_to_auspice [#68]
- Genbank: improved documentation of submission process and slight workflow updates [#51, #71, #75]
- Travis build: github staging / dockstore -- no longer tag non-master branch commits (and instead rely on unversioned branch labels) [#72]
- Travis build: update dxWDL 1.46.4 to 1.47 to reduce dxda file localization instability [#70]
v2.0.21.3
v2.0.21.2
Major changes:
- travis CI changes and cleanups, including new flattened-wdl deploy to github.com/broadinstitute/viral-ngs-staging [#62, #63]
- updates to nextstrain workflows: adds tasks for
augur traits
,augur filter
, and insertsaugur traits
optionally into the main workflow [#59] - metagenomics: addition of new workflows for classify_kraken2, classify_multi (which demonstrates kraken2-based human read depletion) as well as tasks for kraken2-based classification and database building, as well as blastx-based classification. Main workflows such as demux_plus or assemble_denovo have not yet changed default behavior until we get comfortable with the new workflows over various data sets. [#60]