v1.17.0

New:

• The functions to plot coverage in reports.py can now limit graph bounds via --plotXLimits and --plotYLimits
• A function lane_metrics now exists in illumina.py to output cluster density for a run

Fixed:

• Exception now avoided when two sequences do not overlap during multiple alignment/annotation transfer
• Increase picard jvm memory for metagenomics deduplication
• In snakemake metagenomics pipeline, fix the nodupes report incorrectly being dupes

v1.16.0

New:

• illumina.common_barcodes now reports Illumina index names in addition to barcode sequences. Exact matchs are reported, or indices for observed barcodes a Hamming distance of 1 away from known Illumina indices

v1.15.4

Fixed:

• feature notes are now kept even when a feature is clipped
• workaround to tolerate old tbl2asn binaries (they exit with an error code if older than 1 year)
• when a version is not specified for one of the conda-packaged dependencies, any version is now permitted

Changed/Updated:

• in intrahost.vphaser_one_sample the bam file is only sorted if it is not already in coordinate sort order
• update blast 2.2.31 -> 2.6.0
• update diamond 0.8.22 -> 0.8.36
• update picard 2.5.0 -> 2.9.0
• update novoalign 3.06.05 -> 3.07.00

v1.15.3

Updated:

• tbl2asn 25.0 -> 25.3 (and now from bioconda)

v1.15.2

Fixed:

• version detection for tools installed under newer versions of conda

v1.15.1

Changed:

• updated docs (no functional changes to code)

v1.15.0

New:

• Improvements to the pipeline to support single-end reads in additional stages, including the depletion, filtering, and metagenomic steps (blastn, bmtagger, trimmomatic, mvicuna, kraken)
• Scripts added related to auto deployment of viral-ngs conda package
  Updated/changed:
• novoalign 3.04.04 -> 3.06.05

v1.14.0

New:

• read_utils.align_and_fix() now has an option--skipMarkDupes since samtools depth can segfault if if duplicate reads are marked
• reports.align_and_plot_coverage() can now use Novoalign as the aligner rather the default BWA mem

Fixed:

• align_rna memory raised to 36 GB in Snakemake pipeline
• metagenomics.coverage_lca() is now hardened to missing taxids

Changed/Updated:

• For align_rna, "sensitive" option is now disabled by default for BWA mem
• pytest 2.9.1 -> 3.0.5

v1.13.4

Fixed:

• position output for genome features falling oob during annotation is now correct
• in snakemake pipeline:
  • memory requirement has been raised to 7GB for trinity de novo assembly
  • memory requirement for mafft multiple alignment now scales with sample count
  • merge_one_per_sample now sent to the long queue is >1.5GB

v1.13.3

Fixed:

• the vphaser2 step is now hardened against empty input resulting from removal of doubly-mapped reads; in such cases, a blank vphaser output file is now created, allowing the pipeline to continue