Release 6.0

Explanation

This is a major release of SQANTI3. The fundamental changes revolt around the rescue module, with some changes to the order and structure of the QC classification table.

SQANTI3 QC module

There has been an adition of the outputs from TD2 into the classification file, such as the Psauron score, the type of CDS and the protein length. Some columns have been rearranged, to have a more logical order (subcategory goes after structural_category). Internal changes have been done to the classess that handle the isoforms during the workflow, but this will not impact the user experience, only facilitate development and speed.

Several aspects of the classification table have been optimized. The columns of diff_to_TTS and diff_to_TSS have been changed so they show the nucleotide distance of the "spliced" transcripts, and two new columns have been added to show the genomic distance of the TSS/TTS of the isoforms (diff_to_TT/SS_genomic). As well, the ref_legnth and ref_exons columns have been set to NA for non-FSM/ISM transcripts, since they have no reference associated (before the value that was being shown was that of the best reference match)

New plots have been incuded in the report, to take advantage of the information that TD2 outputs, in comparison with GMST.

SQANTI3 Filter module

Minor changes have been introduced here, mainly bugfixes and validation of the categoruies in the rule filter

SQANTI3 Rescue module

The logic behind rescue has been altered, specially when it comes to the rescue-by-mapping pipeline. Now, the best isoform hit will be selected based on the alignment score that minimap2 produces. This will guide the selection of the rescued isoform.

The requantification module has been fully implemented into the code. The couns fo the artifact will be added to the counts of either their best matching isoform (either from the reference or from the long-reads defined transcriptome) or to the Transcirpt Divergency. For more information on the logic, process and results, please visit the wiki, where the pipeline is explained in detail.

SQANTI reads

Quality of life improvements have been introduced, to make clearer for the user the different arguments of SQANTI-reads

What's Changed

• Included json categories check by @pabloati in #491
• Reestructuration of the classes used for classification by @pabloati in #499
• Improved the ORF prediction by @pabloati in #522
• Fix FL column serialization for multi-sample datasets by @Copilot in #535
• Bugfix/filter report by @pabloati in #542
• Fixed quality control report and environment installation and tests action by @pabloati in #563
• Updated abundance file parsing and count processs by @pabloati in #564
• Pythonize rescue by @pabloati in #566
• Rescue testing suite by @pabloati in #567
• Requantification revamp by @pabloati in #571
• Remove pb dependency by @pabloati in #572
• Filter output modified by @pabloati in #573
• Sqanti reads updates by @pabloati in #574
• Conda arm64 bug by @carolinamonzo in #578
• Fixing reads bugs by @carolinamonzo in #577
• Fix example configs by @fmerinocasallo in #582
• Fix TUSCO report crash when gene identifiers are not all-uppercase by @TianYuan-Liu in #588
• Update conda environment specs by @fmerinocasallo in #583
• Refactor transcript parsing and integrate genomic/spliced differences by @pabloati in #590
• New pixi.toml to support Pixi environments by @fmerinocasallo in #584
• Release v6.0 by @pabloati in #591

New Contributors

• @fmerinocasallo made their first contribution in #582

Full Changelog: v5.5.4...v6.0