Skip to content

changelog.md

Latest commit

 

History

History
105 lines (96 loc) · 5.05 KB

changelog.md

File metadata and controls

105 lines (96 loc) · 5.05 KB

Version 2.1

2.1.0

  • Cell-calling is done for each sub-library separately, improving performance for large runs
  • Changes to CellFinder to align more closely with EmptyDrops
    • Cell-barcodes far below the median unique transcript count (medianFraction) are never called cells
  • Beads exposed to a large ambient RNA signal are filtered by default
  • Ultima .cram input supported without fixed filename pattern
  • Short reads after adapter and Poly-A trimming are not longer reported in the library QC reports
    • These reads are included in "Total Sample Reads" in the sample reports
  • Saturation is always calculated on all transcriptome reads
    • Unique and multimappers and both genomes for barnyard samples
  • Barcode naming changes
    • The barcode from the RT plate is called RT and no longer pbc barcode
    • Bead barcode blocks are 01-96, instead of 1A-12H
  • The workflow now checks minimal compute resource requirements after start-up
  • Updated example samplesheet.csv files
    • The index orientation needed for NextSeq 2000 and NovaSeq X is now the default, rather than _revComp

Version 2.0

2.0.1

  • Remove trimming of cDNA (RNA) read to max 82bp length
  • Make QuantumScale RNA the default library type

2.0.0

  • Add support for QuantumScale RNA data
  • Run cutadapt upstream of barcode demux to do adapter trimming
  • Output unaligned BAM files post barcode demux, that are then sent to STAR as input
  • Perform read length filtering during barcode demux
  • Performance optimizations on ScalePlex assignment and reporting
  • For barnyard experiments cell calling is now performed on a species-by-species basis
  • Species specific metrics are now reported in the sample report for barnyard experiments
  • Added sF, gn, and gx tags to BAM file output by STAR
  • Updated read metric definitions, e.g. "Reads mapped to Transcriptome"; see RNA QC reports

Version 1.6

1.6.2

  • Fix erroneous channel dump in inputReads.nf that was triggered by Nextflow 24.10

1.6.1

  • Fix unnecessary parameter warning

1.6.0

  • Add support for ScalePlex library analysis and assignment in tandem with RNA analysis, enabled by the --scaleplex parameter

Version 1.5

  • New optional EmptyDrops-like cell calling method (--cellFinder)
  • Correct up to one N basecall per barcode, like other sequencing errors
  • Rename resource limit parameters (taskMaxMemory, taskMaxCpus, taskMaxTime)
  • Added option to output an anndata object of the cell-count matrix (--annData)
  • Updated optional clustering workflow and report (seurat v5-based)
  • Changed threshold settings (expectedCells, fixedCells, minUTC)
  • Separated pass and flags columns for cell-calling in allCells.csv
  • Make --merge the default for extended-throughput plate runs
  • Fixed mouse background calculation for barnyard runs
  • Compute mitochondrial fraction based on mapped reads
  • Changed barcode rank plot to support CellFinder calls
  • Simplified parallelism strategy
  • Improved task error retry strategy
  • Added more input parameter validations
  • All ScaleBio docker containers moved to AWS ECR

Version 1.4

1.4.1

  • Update datapane version to remove need to access users home dir.
  • Always output sequencing-depth intrapolation results as (rounded) integers
  • Update workflow test dataset to RNA kit v1.1

1.4.0

  • Add support for RNA kit v1.1 (i5-indexed plates)
    • libStructure argument is now required; "libV1.json" or "libV1.1.json"
  • Add support for Extended Throughput kit (multi-plate merging)
  • Add 'reporting workflow' to re-generate outputs from previous alignments
  • Change output file name and directory structure to support multiple per-library and merged outputs
  • Add analysis metadata to sample QC report
  • Remove read2 (RNA) read-length trimming default (previously was 48bp)
  • Changed column label for PCR barcode in allcells.csv (PCR to i7/i5)
  • Fix line-count in the header of the merged, unfiltered STARSolo .mtx files

Version 1.3

1.3.3

  • Updated bcl-convert dependency to version 3.9.3
  • New tiny pipeline test dataset

1.3.2

  • Output gene expression matrix (mtx) from STAR with resolved multimappers included
    • See --starMulti option in nextflow.config
  • Added trimmed reads metric to sample QC report
  • Exclude invalid i7 barcodes during fastq generation (samplesheet.csv)
  • Switch metric file outputs to .csv
  • Option for increased parallelization with --splitFastq
  • New platemap color scheme

Version 1.2

  • Enable Poly-T trimming
    • Reads are now trimmed at Poly-T, in addition to Poly-A stretches, to remove a few antisense mismapping artifacts
  • Metrics re-definition
    • Many QC metric definitions are updated. See qcReport.md
  • Reorganized output (--outDir) directory. See outputs.md
  • Fix conda environment specification
  • Remove Index reads from fastQC
  • Performance improvements

Version 1.1

  • Enable Poly-A trimming
    • By default all cDNA reads are now trimmed at the first occurrence of a 8bp Poly-A stretch
    • This avoids mismapping of RT primer sequences

Version 1.0

Initial Release