Skip to content

Releases: Integrative-Transcriptomics/MUSIAL

MUSIAL-v2.3

28 Mar 14:25
@s-t-h s-t-h
v2.3
f8940b7
Compare
Choose a tag to compare
MUSIAL-v2.3 Latest
Latest

v.2.3.0

  • Added container classes to increase abstraction.
  • Increased efficiency of computations and memory usage.
  • All variant calls are now stored in the build output, including reference. This reduces reference bias by rejecting calls.
  • Simplified input parameters (call quality is no longer considered).
  • Removed deprecated/redundant code.

v2.3.1 (Minor Update, 21.05.2024)

  • Improved compatibility with bcftools called variants.
  • Switched to gzip for compression to remove OS dependent errors.
  • Removed deprecated/redundant code.

v2.3.2 (Minor Update, 03.06.2024)

  • Added option to the build task to specify temporary working directory used by SnpEff.
  • Rejected variants are now indicated by a ? symbol instead of !.

v2.3.3 (Minor Update, 05.06.2024)

  • The option -u of the build task can now be used to write uncompressed storage files.
  • Improved resolution of complex InDels.
  • Removed skipping of non-haploid samples, still, all input files will be considered as haploid.

v2.3.4 (Minor Update, 14.06.2024)

  • Bugfix for aminoacid variant inference.

v2.3.5 (Minor Update, 26.06.2024)

  • Bugfix for aminoacid and nucleotide variant inference, i.e., in some scenarios InDels were treated as reference calls and ignored in downstream processing.

v2.3.6 (Minor Update, 28.06.2024)

  • Bugfix for processing deletions that exceed feature lengths during proteoform inference.

v2.3.7 (Minor Update, 10.07.2024)

  • Add option to exclude, in addition to positions, explicit variants from the analysis (to tackle reference errors).

v2.3.8 (Minor Update, 26.07.2024)

  • Reference allele/proteoform is now stored independent of occurrence in samples.
  • Separator symbol ";" replaced by ",".
  • Extended nucleotide-variant storage logic to comprise variants that pass the filter criteria but are not the most frequent in one sample:
    • Nucleotide variants stored for a feature are, in this sense, either (i) the most frequent allele passing filter criteria, (ii) the most frequent allele failing filter criteria, i.e., an ambiguous call, or (iii) not the most frequent allele, but passing filter criteria, i.e., non-primary variants.
    • This is reflected by a primary attribute in the variant information.
    • Sequence and table export still only consider primary/most frequent variants.
    • Primary variants (derived of the most frequent allele of a call) that do not pass filter criteria are considered ambiguous variants and are now stored as SNVs with content N instead of Ns in the length of possible InDels to reduce bias towards actual variants.
    • Actual variant content is now stored for ambiguous variants.

v2.3.9 (Minor Update, 16.08.2024)

  • Bug fixes:
    • The .fai index files for the reference sequence are now overwritten on each run to avoid supposed errors due to changes in the file contents.
    • All do-while loops have been replaced by while loops to avoid errors due to empty iterators; this only happened in the case that no reference allele was present in the sample input.
    • Fixed an error in the logical operation for filtering variants.
  • SnpEff now annotates ambiguous variants, i.e. filtered variants, with respect to their actual alternative nucleotide content and extended the annotation to variants that are not on coding genes.

v2.3.10 (Minor Update, 10.09.2024)

  • SnpEff has been updated to the latest version and minor parameter changes have been made to increase the efficiency of the SnpEff runtime.
Assets 3
Loading