xPore v2.1

Summary

xPore v2.1 has some minor changes from the previous version, which include the following.

• All k-mers containing ‘N’ will be discarded in both options i.e. with and without --genome in xpore dataprep.
• New GFF option: to map the transcriptomic to genomic coordinates (when using --genome in xpore dataprep), xPore requires both FASTA for sequence information and either GTF or GFF for coordinate information. However, GTF is the recommended option. If GFF is the only file available, please note that the GFF file works with GENCODE or ENSEMBL FASTA files, but not UCSC FASTA files. We plan to remove the requirement of FASTA files in a future release.
• In xpore dataprep, the option names have been changed as follows.
  • --gtf_path_or_url to --gtf_or_gff
  • --transcript_fasta_paths_or_urls to --transcript_fasta
• The link to our demo dataset in zenodo was corrected.
• The documentation was updated to include these changes.

What's Changed

• gtf option, fastdataprep, and dataprep intermediate file removal by @yuukiiwa in #48
• Update README.md by @jonathangoeke in #49
• remove pyensembl requirement and solve Arabidopsis bug by @yuukiiwa in #57
• Update README.md by @jonathangoeke in #51
• replace if statement with KeyError in g2t mapping, add post-processing option, and add xpore wrapper by @yuukiiwa in #58
• Update README.md by @jonathangoeke in #66
• Update help.rst by @jonathangoeke in #67
• readGTF function was modified to adapt to non-ensembl GTF input by @obenno in #87
• add assert "NNNNN" for process_tx and solve bug with stringtie gtf by @yuukiiwa in #96
• Fix quickstart doc by @hendrikweisser in #103
• Add GFF annotation option and change fasta and gtf input flags names by @yuukiiwa in #90
• update readthedocs with --gtf_or_gff specifics by @yuukiiwa in #105

New Contributors

• @yuukiiwa made their first contribution in #48
• @jonathangoeke made their first contribution in #49
• @obenno made their first contribution in #87
• @hendrikweisser made their first contribution in #103

Full Changelog: v0.5.6...v2.1

xPore v2.0

Summary

xPore v2.0 is a major release that implements the following changes:

• xPore now has 3 main running modes that replace the scripts in earlier versions:
  • xpore dataprep
  • xpore diffmod
  • xpore postprocessing
• Checking the version is now supported by xpore -v
• Changes in the xpore dataprep command as follows:
  • the pyensembl package is no longer required and supported to map transcriptome to genome coordinates. Please use the --genome flag and provide the GTF file (--gtf_path_or_url) and transcriptome FASTA file (--transcript_fasta_paths_or_urls) if you would like to work with genomic coordinates.
  • The option --summary is no longer needed and was removed.
  • xpore dataprep will no longer produce 'N' in k-mers.
• The documentation was updated to include these changes.
  xPore is still compatible with dataprep files generated with xPore v1.0.

What's Changed

• gtf option, fastdataprep, and dataprep intermediate file removal by @yuukiiwa in #48
• Update README.md by @jonathangoeke in #49
• remove pyensembl requirement and solve Arabidopsis bug by @yuukiiwa in #57
• replace if statement with KeyError in g2t mapping, add post-processing option, and add xpore wrapper by @yuukiiwa in #58

Full Changelog: v0.5.6...v2.0

xPore v1.0

Summary

• xpore-dataprep now returns a single data file (data.json) along with 2 small index files, a read count information file, and a log file.
• For 1 million reads, running xpore-dataprep with 1 CPU takes approximately 5 hours with 5 GB of RAM, and the uncompressed data.json file is 4GB (compared to ~150GB for fast5 files or ~200GB for Nanopolish output files). The json file is a text file and can be efficiently compressed to ~1GB.

What's Changed

• gtf option, fastdataprep, and dataprep intermediate file removal by @yuukiiwa in #48
• Update README.md by @jonathangoeke in #49

Full Changelog: v0.5.6...v1.0

xPore v0.5.6

Small bugs fixed.

xPore v0.5.5

The read-level modification rates can be saved for further analysis using --save_models option.

xPore v0.5.4

• #24 solved.
• Small bugs fixed

xpore v0.5.3

• #18 Transcript IDs are kept in full format -- no more trimming the transcript version after dot (.). So, other species that are not in Ensembl yet can be at least now run using the transcriptome mode in xpore-dataprep.
• #19 xpore-dataprep can read the compressed eventalign file (.gz).

xpore v0.5.2

• #18 Transcript IDs are kept in full format -- no more trimming the transcript version after dot (.). So, other species that are not in Ensembl yet can be at least now run using the transcriptome mode in xpore-dataprep.
• #19 xpore-dataprep can read the compressed eventalign file (.gz).

--ERROR--

xpore v0.5.1

Minor update from v0.5.0

• Species known by ensembl other than human can be specified during the xpore-dataprep process.