v1.3.5

• Add ability to provide SV VCF column in GroupedAnalysisAmpSuite.py input file for each sample individually.
• Add some warnings for grouped analysis if input file is formatted incorrectly.
• Expose --pair_support_min and --foldback_pair_support_min arguments from AA to AmpliconSuite-pipeline.py. These are exposed in the Docker & Singularity runscripts too.
• Release with AA v1.5.r0.

v1.3.4

• clean up temporary reference fasta when matched normal given
• fix logging of AA version when python2 used
• clear previous run logfile when launching a rerun (previously would append to end of previous log)
• bugfix to GroupedAnalysis when multiple normal samples given

v1.3.3

AmpliconSuite-pipeline 1.3.3

• Accompanies AmpliconClassifier 1.2.0 and AmpliconArchitect 1.4.r2 updates
• fix bug in GroupedAnalysisAmpSuite.py where not all samples analyzed if number of threads is smaller
• add checking of exit codes from subprocesses in GroupedAnalysisAmpSuite.py
• require user to provide location of .sif file for Singularity usage.
• lazy loading of some libraries so that --download_repo can be used without installing any dependencies
• improvements in logging

v1.3.2

AmpliconSuite-pipeline 1.3.2

• Accompanies AC v1.1.4 update
• reduce_fasta.py is now integrated as a library file, not a script
• improved error messages about mosek license
• rename intermediate file pre_filtered.bed to unfiltered_gains.bed and put in cnvkit directory to avoid confusion about its contents.
• Packaged versions do not encounter GroupedAnalysisAmpSuite.py errors during feature similarity calculation and when a matched normal is given as input.

v1.3.1

• Remove bloat from container images (Docker 8.7GB -> 2.2GB. Singularity 6.0GB -> 636MB)
  • This was due to unnecessary, massive pytorch GPU packages being included with CNVkit's dependencies by default.
• Remove bloat from standalone installer script (pytorch GPU packages)
• Add bedtools as a dependency
• GroupedAnalysisAmpSuite.py will automatically compute feature similarity scores.
• GroupedAnalysisAmpSuite.py bugfix when CNV seeds are already provided in input table.
• GroupedAnalysisAmpSuite.py detect improperly formatted input table.
• Automatically detect samtools version (removes --use_old_samtools)
• Warn user if alt contigs in bam header.
• Fix logging of AA version in run_metadata.json.
• Better protection for bad inputs (e.g. user gave same fastq twice, bam file with spaces in name)
• Fix no amp samples not getting ref genome version in run_metadata.json
• Update Rscript package installation in containers.
• Bugfix to manual specification of Singularity image file location

v1.2.2

• Bugfix to GroupedAnalysisAmpSuite.py when matched normal not included
• Distributable packages to include v1.3.r8 of AmpliconArchitect
• Distributable packages to include v1.1.2 of AmpliconClassifier

v1.2.1

• Bugfix in AmpliconClassifier

AmpliconSuite-pipeline version incremented for re-packaging.

v1.2.0

• adds --download_repo option so users can obtain data repo files using AmpliconSuite-pipeline.py
• update install.sh
• conda recipe update for AC 1.1.0
• documentation updates and logging tweaks

v1.1.3

• Bugfixes to handling of CNV seeds file when using GroupedAnalysisAmpSuite.py
• Update for AA v1.3.r7

v1.1.2

• debug linkage with AmpliconClassifier for conda-installed versions of AmpliconClassifier
• minor changes to docs and logging
• add manifest.in