With Montgomery

[kclem]

What's Changed

This release improves parallel processing performance and increases flexibility in input file headers and output file naming conventions.

Major updates

• Parallelization updates
  • Don't start pool when only using single thread by @Colelyman in Fix multiprocessing error, don't start pool when only using single thread #302
  • Parallel plotting fix by @Colelyman and @kclem in 546446e and Parallel plotting fix #286
  • Raise exceptions from within futures in plot_pool in a439f09
  • Enable CRISPRessoPooled multiprocessing when os allows multi-thread file append inhttps://github.com/pinellolab/CRISPResso2/commit/ebb016dff46c280dce8c3c09e8ac0e0cc25d4d74
• Allow multiple overlapping sgRNA matches in reference (previous behavior was to only search for non-overlapping sgRNA sites in the reference sequence in 32e1e97
• Assert correct input fastq file format in 7248ba8

Minor updates:

• Sort pandas dataframes by # of reads and sequences so that the order is consistent for testing by @Snicker7 and @Colelyman in Sort pandas dataframes by # of reads and sequences so that the order is consistent #316
• Update base_editor parameters in README and add Plot Harness by @Colelyman in Update base_editor parameters in README and add Plot Harness #301
• Add verbosity argument to CRISPRessoAggregate (Insertions longer than the remaining reference sequence are missed #18) fixes Issue with args in CRISPRessoAggregate #306 by @Colelyman in Add verbosity argument to CRISPRessoAggregate (#18) fixes #306 #307
• Clarify CRISPRessoWGS intended use by @Colelyman in Clarify CRISPRessoWGS intended use #303
• Update plotCustomAllelePlot.py script for custom plot script #292 by @kclem in Update plotCustomAllelePlot.py script for #292 #293
• Case-insensitive headers accepted in CRISPRessoPooled e577318
• Fix multiprocessing lambda pickling by @Colelyman in Fix multiprocessing lambda pickling #311
• Allow dashes in filenames in 712eb2a

Full Changelog: v2.2.12...v2.2.13

---

This discussion was created from the release With Montgomery.

[Jdbeck66]

Is there any update on the release date for 2.3 with the change to Fastp? Also curious about participation in this project?

[Colelyman]

Thank for your comment. We don't have an expected release date for 2.3 because we still need to verify that everything works as expected. If you are anxious to use Fastp, the branch is at https://github.com/edilytics/CRISPResso2/tree/fastp (we would welcome your input!).

What questions do you have about participating? Do you have an idea for a feature?

[Jdbeck66]

Thanks for the info about 2.3.

In regards to participation, I'd like to implement a couple of useful ideas found in the CRISPAltRations paper. Specifically, I'd like to be able to calculate the quantification window size using treated and untreated samples as inputs.

Also, I'd like to implement a position specific incentive vector that aligns with the cut profile of a specific Cas protein. Currently, I can only make a macro decision about open and extension incentives. I'd like to pass in a vector that is specific and positionally different surrounding the cut site.

Finally, there is interest in my company about identifying background noise from untreated samples and using that to filter/adjust/mark treated samples near that thresh hold. Thanks for your consideration.