CHARLIE v0.13.0

CHARLIE 0.13.0

• Migrate runtime activation to a mamba-safe flow to avoid shared-environment activation failures. (#149, @kopardev)
• Host-only circRNA discovery is now supported. Additives (e.g. ERCC) and viruses are now fully optional — omitting them runs the pipeline in host-only mode, a long-standing user request. Make host/additives/viruses config placeholders envsubst-compatible and default additives to blank when omitted. (#149, @kopardev)
• Replace hard-coded CCBR base paths under /gpfs with /vf for current filesystem layout compatibility. (#149, @kopardev)
• Update Biowulf partition handling by removing deprecated ccr usage and forcing/defaulting to norm. (#149, @kopardev)
• Align onsuccess/onerror completion hooks with CARLISLE behavior while preserving CHARLIE's shared hook implementation, including stricter shell setup and completion-log checks. (#153, @kopardev)
• Auto-detect $SIFCACHE (set by module load ccbrpipeliner) and use it as the singularity cache and --singularity-prefix so Snakemake finds pre-downloaded SIF files without pulling from Docker Hub. (#154, @kopardev)
• Print a configuration summary block (workdir, snakefile, config, singularity cache, binds, platform) to the terminal before each pipeline run. (#154, @kopardev)
• Fix set_singularity_binds not being called in unlock(), dryrun(), and touch(), which left SINGULARITY_BINDS empty. (#154, @kopardev)
• Smarter singularity bind path detection: walk up parents of WORKDIR and PIPELINE_HOME to find /data/<name> symlinks and include both the symlink and real path; filter Python-derived paths that are ancestors (too broad) or descendants (too specific) of the bash-derived explicit paths. (#154, @kopardev)

CHARLIE 0.12.0

• Now depends on ccbr_tools v0.4 for updated jobby & spooker utilities. (#142, @kelly-sovacool)

CHARLIE 0.11.2

• Minor documentation updates. (#138, @kelly-sovacool)

CHARLIE 0.11.1

• CHARLIE was falsely throwing a file permissions error for tempdir values containing bash variables. (#118, @kelly-sovacool)
• Singularity bind paths were not being set properly. (#119, @kelly-sovacool)
• Update docker containers to set $PYTHONPATH. (#119, #125, @kelly-sovacool)
  • Otherwise, this environment variable can be carried over and cause package conflicts when singularity is not run with -C.
  • Also use python -E to ensure the $PYTHONPATH is not carried over. (#129, @kelly-sovacool)
• Fix reconfig to correctly replace variables in the config file. (#121, @kelly-sovacool)
• Prevent using excessive memory when copying reference files. (#126, @kelly-sovacool)
• Fix missing output files due to file system latency and use real (absolute) paths where possible. (#130, @kelly-sovacool)
• Update documentation to reflect biowulf usage and improved test dataset. (#132, @kelly-sovacool)

CHARLIE 0.11.0

• Major updates to convert CHARLIE from a biowulf-specific to a platform-agnostic pipeline (#102, @kelly-sovacool):
  • All rules now use containers instead of envmodules.
  • Default config and cluster config files are provided for use on biowulf and FRCE.
  • New entry TEMPDIR in the config file sets the temporary directory location for rules that require transient storage.
  • New --singcache argument to provide a singularity cache dir location. The singularity cache dir is automatically set inside /data/$USER/ or $WORKDIR/ if --singcache is not provided.
• Minor documentation improvements. (#114, @kelly-sovacool)

v0.10.1

• strand are reported together, strand from all callers are reported,
• both + and - flanking sites are reported,
• rev-comp function updated,
• updated versions of tools to match available tools on BIOWULF.

v0.9.0

Significant upgrades since the last release:

• updates to wrapper script, many new arguments/options added
• new per-sample counts table format
• new all-sample master counts matrix with min-nreads filtering and ntools column to show number of tools supporting the circRNA call
• new version of Snakemake
• cluster_status script added for forced completion of pipeline upon TIMEOUTs
• updated flowchart from lucid charts
• added circRNAfinder, find_circ, circExplorer2_bwa and other tools
• optimized execution and resource requirements
• updated viral annotations (Thanks Sara!)
• new method to extract linear counts, create linear BAMs using circExplorer2 outputs
• new job reporting using jobby and its derivatives
• separated creation of BWA and BOWTIE2 index from creation of STAR index to speed things up
• parallelized find_circ
• better cleanup (eg. deleting _STARgenome folders, etc.) for much smaller digital footprint
• multitude of comments throughout the snakefiles including listing of output file column descriptions
• preliminary GH actions added

v0.7.0

• 5 circRNA callers
• all-sample counts matrix with annotations

v0.6.9

• Optimized pysam scripts
• fixed premature completion of singularity rules

v0.6.5

• updated config.yaml to use the latest HSV-1 annotations received from Sarah (050421)