Raw code usage:

• LIBRA fine-tuning code (LIBRA_fine_tune_code)

• Use paired-omics arrays as input for fine_tune_libra.py

  • Use "fine_tune_libra.py" for LIBRA models training in parallel using desired grid of hyperparameters.

    Outputs: Different outputs generated during the training will be stored in the working directory.

• PPJI metric computation over all fine-tune models:

  • Use supp_code.R stored at R code employed on LIBRA manuscript.