#!/bin/bash -euo pipefail
python -u /usr/local/src/CTAT-SPLICING/STAR_to_cancer_introns.py \
    --SJ_tab_file GM24385_0001_RNA_0001_23H5VFLT4_s16.SJ.out.tab \
    --chimJ_file GM24385_0001_RNA_0001_23H5VFLT4_s16.Chimeric.out.junction \
    --bam_file GM24385_0001_RNA_0001_23H5VFLT4_s16.Aligned.sortedByCoord.out.bam \
    --output_prefix GM24385_0001_RNA_0001_23H5VFLT4_s16 \
    --ctat_genome_lib ctat_genome_lib_build_dir \
    --vis --sample_name GM24385_0001_RNA_0001_23H5VFLT4_s16

# Sort output files to ensure consistent ordering across runs while preserving headers
if [ -f GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns ]; then
    # Extract header (first line) and sort data lines by fourth column (uniq_mapped) descending, then by first column (intron)
    head -n 1 GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns > GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns.tmp
    tail -n +2 GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns | LC_ALL=C sort -k4,4nr -k1,1 >> GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns.tmp
    mv GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns.tmp GM24385_0001_RNA_0001_23H5VFLT4_s16.cancer.introns
fi

cat <<-END_VERSIONS > versions.yml
"NFCORE_RNAFUSION:RNAFUSION:CTATSPLICING_STARTOCANCERINTRONS":
    ctat-splicing: 0.0.3
END_VERSIONS