proposed v1.9.1 updates

[gpertea]

Update of internal documentation and examples

• renamed the covComb_tx_deg rse object to rse_tx in all code comments and test code or other documentation mentioning it; data/ and data-raw/ rda files and code were adjusted accordingly.
  • motivation: the previous object name was confusing as it was the same with one of the degradation experiments RSEs used to build the internal transcripts list, although the comments and examples in the code were suggesting that it is an unrelated user-provided example RSE to apply qsvaR functions on it.

Improved usage of the package with custom degradation data/model

• better support for a user provided list of transcripts (sig_transcripts) to use for qsvaR correction
  • getDegTx() and qSVA() no longer need to provide the type parameter if sig_transcripts is used (type is then considered to be "custom")
• enhanced DEqual() to include additional parameters to customize the plot (deg_tstats, show.legend, show.cor, font.size, cor.size, cor.label) and improved input validation and error handling:
  • the deg_tstats parameter allows the user to provide the t-stats data from a custom degradation model
  • this change and the above adjustment of the custom sig_transcripts option, allow full use of the qsvaR package with a custom degradation model and list of transcripts, totally independent of the original built-in degradation experiment data.

Internal utility function changes

• the utility function check_tx_names was removed in favor of a combination of 2 new utility functions: normalize_tx_names and which_tx_names, which more efficiently help with the internal matching and mapping of Ensembl/Gencode IDs, removing the versioning properly etc.
  • motivation: the previous approach failed to map such versioned Ensembl IDs in some cases, resulting in a lower number of transcripts being considered for getDegTx()

Minor version update

• updated package version to 1.9.1