SQANTI-SIM 0.2.0

🚀 New Features and Major Changes

• Implemented the simulation of CAGE-seq orthogonal data (--CAGE) when running SQANTI-SIM in sample mode.
• Added PacBio long read simulation using PBSIM3 (--pbsim).
• Added the full-sim mode, which executes the classif, design and sim modules, allowing users to run the complete simulation with just one simple instruction.
• Provided pre-trained models and characterized datasets of the WTC11 human cell line.
• Updated pre-trained NanoSim models with WTC11 human cell line dRNA and cDNA data. Additionally, users have the option to utilize their specific pre-trained NanoSim models using the --nanosim_model parameter.
• Enhanced the --diff_exp parameter by deprecating --low_prob and --high_prob. Now, --diff_exp is a numeric value that allows users to adjust with a single parameter the bias of known and novel expression levels.
• Updated SQANTI-SIM evaluation report style.

🐛 Bug Fixes and Minor Changes

• Added the option --expr_file to provide an expression distribution file to sqanti-sim.py design sample to avoid mapping the same reads in different executions.
• When using the --long_reads option in sqanti-sim.py design sample mode, SQANTI-SIM now generates an expression file named "sqanti-sim_expression.tsv." This file can be reused in future executions using --expr_file.
• Added a default number of novel transcripts to simulate when the user does not specify any.
• Added additional example files, providing users with practical examples.
• Updated the classification algorithm to the latest SQANTI3 release, version 5.1.2, ensuring up-to-date transcript classification results.
• Updated the sqanti-sim.py design sample random expression sampling by implementing inverse transform sampling from ECDFs.
• Fixed Partial True Positives (PTP) and Positive Detection Rate (PDR) metrics to consider only non-redundant transcript models.
• Fixed conda environment installation issues related to bcbio-gff and updated dependencies.
• Fixed SQANTI-SIM Report generating empty tables when not all structural categories were detected.