XPRESSpipe-v0.4.0

• Introduced rRNA depletion during alignment step (previously could only do so during the quantification step)
• Expanded periodicity analysis to cover more holistic P-site analysis (report codon usage stats). The submodule previously called periodicity is now called by p_sites
• Allow for setting upper limit threshold for read length during pre-processing reads and P-site analysis (previously only a lower limit was available)
• All documentation associated with this changes has been updated.

XPRESSpipe-v0.3.1

• Fix BAM file threshold for metagene and geneCoverage to avoid OOM errors
• Turn off BAM file threshold for counting (low memory footprint, so can use all cores available)
• Import openssl library manually in Rperiodicity -- occasionally had trouble finding the library on its own and would error

XPRESSpipe-v0.3.0

• Transfers R dependency installs to Anaconda environment load
• Modified fastq and bam memory factor to optimize resources
• Rebuilt read distribution module with JuliaLang for super memory efficiency during parallelization
• Fixed issue where one-exon genes would not display feature annotations in geneCoverage modules
• Made matplotlib backend calls flexible for HPC usage
• Made directory checks more thorough
• Fixed a potential off-by-one issue with GTF truncator
• Updated appropriate tests
• Updates to documentation
• Added code of conduct and contributions information

XPRESSpipe-v0.2.4b0

Updated journal release version

• Fixed issues with using polyX adaptors
• Allowed more multi-threading during post-processing of aligned reads to use resources more efficiently
• Added integrated pipeline tests for Travis CI build to assess pipeline integrity each push
• Updated install walkthrough video

XPRESSpipe-v0.2.3b0

• Implemented minor changes to let code continue to run if hits error (for example, continue QC tests if generic gene name doesn't have a match in organism of interest)
• Updated some tests and docs

XPRESSpipe-v0.2.2b0

Journal Submission Release

Changes to metagene profile setup

• Before there were big spikes due to skewing from super-expressors (i.e., 10x reads as transcripts with next most reads)
• a4b9388 and 289d056 summarize logic behind methodology

Other changes

• More personable command builder guide
• Updates to documentation
• Looser parallelism threshold for more BAM files at once since it appears to be handling the memory pressure fine right now

XPRESSpipe-v0.2.1b0

• Fixes an issue where gene name search for geneCoverage module was not strict enough and would fit if string was found anywhere in gene name. Now requires a strict match for carry through in the module

XPRESSpipe-v0.2.0b0

bioRXiv release

• doi: https://doi.org/10.1101/704320

Notes:

• Most upcoming changes will be documentation and developing more tests
• readDistribution is RAM hungry and unstable. Removed from pipeline runs for now, but still available as a separate module. Users are encouraged to refer to MultiQC output in the meantime to determine read distributions
• Metagene appears to be stable but will look into the spikes that consistently appear in plots

XPRESSpipe-v0.1.4b2

• Updated modifyGTF to parse for Ensembl canonical transcripts and all riboseq-related modules look for CDS instead of exon (i.e. truncator, periodicity plotting)
• Implemented riboseq pipeline test with checking

XPRESSpipe-v0.1.3b2

• Configured setup to install cufflinks before setup execution and to port cufflinks binary to site packages to be accessible during run-time
• Will automatically check for conda, install if not available, and then install conda dependencies