v0.2.11

• Includes 10x-compatible count matrix in the final outputs.
• Uses --force-glacier-transfer to force download operations on restored Glacier objects.

v0.2.10

• Generates a Cell Ranger compatible count matrix.
• Due to the known non-platform agnostic glob behavior, it is possible that L001_R1 is merged with L002_R2 (not L001_R2). To avoid this problem, we first sort genomic_fastq and barcode_fastq before merging. This is a temporary hacky solution.

v0.2.9

• Fixed the bug that outputs slightly less number of UMI correction data. This fix should give a little better result for RNA Velocity.

v0.2.8

• Set min-poly-t to 0 for 10x v2 and 10x v3.
• Should run with the new AMI that contains STAR 2.7.6a.

v0.2.7

• Increased memory requirements for RMT correction.
• Uses r5.4xlarge as default instance type.
• Logs the running time before uploading the log file.

v0.2.6

• SEQC run
  • Refactored RMT correction module to use Dask and Numba to reduce the running time by 3-4 hours.
  • Uses np.int64 to store UMI (not np.int32).
  • Generates CB/UMI correction files (info needed for tagging BAM)
  • Fixed the bug that incorrectly marks some cells as RMT error.
  • Replaced Deprecated Pandas functions (e.g. as_matrix)
  • Uses Pickle protocol=4 to serialize a byte object larger than 4 GiB
  • Now requires pandas>=1.0.4, statsmodels=0.11.1, scipy>=1.5.1.
  • Uses PhenoGraph 1.5.7.
  • Writes software versions in the log file (SEQC, STAR, and samtools)
  • Log running time.
  • Terminates gracefully when running locally.
  • Terminates gracefully when error occurred while running locally.
  • Handles output_prefix properly.
• SEQC index
  • Supports --local, --emsemble-release, and --read-length.
  • Handles when genome FASTA is gzipped.
  • Changed the default valid biotypes to protein_coding, lincRNA, antisense, IG_V_gene, IG_D_gene, IG_J_gene, IG_C_gene, TR_V_gene, TR_D_gene, TR_J_gene, TR_C_gene.
  • Preserve STAR log file (Log.out).
  • Improves logging.
  • Fixed to respect the --folder and --valid-biotypes argument.

v0.2.5

• To handle snRNA-seq data and maintain the backward compatibility, --filter-mode is introduced.
• Allows to override --max_insert_size if the filter mode is snRNA-seq.
• Sets figure facecolor to white by default in the analysis notebook template (go dark theme!)

v0.2.4

• Change the default EC2 instance type to r5.2xlarge
• Update the installation steps
• Fix MAGIC & PhenoGraph installation
• Install MAGIC v0.1.1 and PhenoGraph v1.5.2 from the dpeerlab repository
• Use magic.MAGIC_core in the notebook

v0.2.3-alpha.5

• Support 10x v3 chemistry
• Add additional command-line arguments:
  • --ami-id: controls which AMI to use
  • --user-tags: adds custom tags to EC2 instance
  • --remote-update: installs the local SEQC package onto EC2 instance
  • --terminate and --no-terminate: controls whether or not EC2 instance will be terminated upon completion
• Set version to v0.2.3

v0.2.2

• In-drop v5 support