Spring Roll

Major updates:

• New 'fitnessDropoutPseudocount' option to add pseudocounts to output replicates with dropout i.e. variants present in input but absent from output (default:0). This is a 'workaround' for these (typically deleterious) variants whose fitness scores would otherwise be undefined or entirely missing from DiMSum output files. Note that pseudocounts are added during stage 5 and after error modeling to ensure that pseudocounts do not contribute to the error model fit.

Wonton

Major updates:

• New 'vsearchMaxQual' option to support maximum base qualities >41 e.g. as provided by the Element AVITI sequencer

Dumpling soup

Major updates:

• Performance improvements and approximately 10-fold reduced volume of data input and output (intermediate files)
• Support for any number of biological replicates (previously limited to less than 10)

Haam Seui Gok

Major updates:

• New command-line argument "--synonymSequencePath" in order to supply additional reference variants (other than the wild-type) for which synonymous substitution variant fitness and error estimates are required
• "all_variants" R object now excludes synonymous variants for coding sequences (i.e. only WT and missense mutations); these are instead included in the "synonymous" R object (previously "silent" R object)
• "fitness_doubles.txt" output file now contains all doubles (not just those where both constituent singles were quantified)

Minor updates:

• Synonymous variant fitness and error estimates are now provided when "--mixedSubstitutions" option set to TRUE
• Synonymous variants saved to "fitness_synonymous.txt"( instead of "fitness_silent.txt"), WT nucleotide variant is included as well as all sample read counts
• "fitness_wildtype.txt" now contains all sample read counts
• Fixed bug in "dimsum__concatenate_reads"
• Updated checks on Variant Count and Barcode Identity files

Taro dumpling

Minor updates:

• Fixed issue #6 related to negative scaling factors in replicate normalisation parameters ("dimsum__error_model.R")
• Fixed bug in "dimsum__sample_count_distribution.R"
• Allow no more than 9 biological replicates to be specified

Shumai

Major updates:

• Fixed bug when translating sequences using Biostrings::translate which by default assumes that the first codon in a DNA sequence is an initiation codon (fixed with no.init.codon=T)

Minor updates:

• Improved .html report
• If supplied, the Variant Count File is now checked for duplicated nucleotide sequences

Guotie

Major updates:

• Infer growth rates from read counts if columns including "cell_density" (optical density or similar) and "selection_time" (in hours) supplied in Experimental Design File
• Requires VSEARCH >=v2.17 (improved algorithm and now allows merging of sequences with overlaps as short as 5 bp: https://github.com/torognes/vsearch/releases/tag/v2.17.0)

Minor updates:

• Fixed bug in dimsum__merge_report

Xiaolongbao

Minor updates:

• Updated "dimsum__merge_fitness.R" to support arbitrary library complexities
• Fixed issue #2 which caused an error in Stage 1
• Other minor bug fixes and improvements

Chive dumpling

Minor updates:

• Support for adaptive minimum input/output read count thresholds
• Optional trimming of either/both 5' and 3' constant regions

Teochew dumpling

Major updates:

• Support for indel variants (see "indels" command-line arguments)
• Removed "vsearchMinlen" option ("cutadaptMinLength" option now serves as only filter on read length)
• Fixed bug in "dimsum__aggregate_AA_variants_fitness.R" (in some cases aggregated variant fitness and sigma could be set to NA even if other synonyms were not NA)

Minor updates:

• Improvements to output files and report.html
• Other minor bug fixes and improvements