name: feat_annotation
version: 0.0.2
output_asset_class: feat_annotation
description:
  Combined genomic feature annotation created using an Ensembl GTF annotation
  asset and an Ensembl regulatory build annotation asset
input_files: {}
input_params: {}
input_assets:
  ensembl_gtf:
    asset_class: gtf
    default: ensembl_gtf
    description: Annotation file in Gene Transfer Format (GTF) from Ensembl
  ensembl_rb:
    asset_class: ensembl_rb
    default: ensembl_rb
    description: Regulatory annotation file in General Feature Format (GTF) from
      Ensembl
docker_image: null
command_templates:
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_gtf"].seek_keys_dict["gtf"]}} | awk '$3=="exon"' | grep -v 'pseudogene' | awk -v OFS='\t' '{print "chr"$1, $4-1, $5, "Exon", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3n -u > {{values.output_folder}}/{{values.genome_digest}}_exons.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_gtf"].seek_keys_dict["gtf"]}} | awk '$3=="exon"' | grep -v 'pseudogene' | awk -v OFS='\t' '{ split($20, a, "\""); print "chr"$1, $4-1, $5, a[2], $6, $7}' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3n -u | awk 'seen[$4]++ && seen[$4] > 1' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3nr | env LC_COLLATE=C sort -k1,1 -k2,2n -u | env LC_COLLATE=C sort -k1,1 -k3,3n -u | awk -v OFS='\t' '{if($4==prev4){new2=prev3+1;} {prev4=$4; prev3=$3; print $1, new2, $2-1, "Intron", $5, $6}}' | awk -F'\t' '$2' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -u > {{values.output_folder}}/{{values.genome_digest}}_introns.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_gtf"].seek_keys_dict["gtf"]}} | awk '$3=="three_prime_utr"' | grep -v 'pseudogene' | awk -v OFS='\t' '{print "chr"$1, $4-1, $5, "3\047 UTR", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -u > {{values.output_folder}}/{{values.genome_digest}}_3utr.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_gtf"].seek_keys_dict["gtf"]}} | awk '$3=="five_prime_utr"' | grep -v 'pseudogene' | awk -v OFS='\t' '{print "chr"$1, $4-1, $5, "5\047 UTR", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -u > {{values.output_folder}}/{{values.genome_digest}}_5utr.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_rb"].seek_keys_dict["ensembl_rb"]}} | awk '$3=="promoter"' | awk -v OFS='\t' '{print "chr"$1, $4, $5, "Promoter", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3n -u > {{values.output_folder}}/{{values.genome_digest}}_promoter.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_rb"].seek_keys_dict["ensembl_rb"]}} | awk '$3=="promoter_flanking_region"' | awk -v OFS='\t' '{print "chr"$1, $4, $5, "Promoter Flanking Region", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3n -u > {{values.output_folder}}/{{values.genome_digest}}_promoter_flanking.bed
  - gzip -dcf {{values.genome_folder}}/{{values.assets["ensembl_rb"].seek_keys_dict["ensembl_rb"]}} | awk '$3=="enhancer"' | awk -v OFS='\t' '{print "chr"$1, $4, $5, "Enhancer", $6, $7}' | awk '$2<$3' | env LC_COLLATE=C sort -k1,1 -k2,2n -k3,3n -u > {{values.output_folder}}/{{values.genome_digest}}_enhancer.bed
  - cat {{values.output_folder}}/{{values.genome_digest}}_enhancer.bed {{values.output_folder}}/{{values.genome_digest}}_promoter.bed {{values.output_folder}}/{{values.genome_digest}}_promoter_flanking.bed {{values.output_folder}}/{{values.genome_digest}}_5utr.bed {{values.output_folder}}/{{values.genome_digest}}_3utr.bed {{values.output_folder}}/{{values.genome_digest}}_exons.bed {{values.output_folder}}/{{values.genome_digest}}_introns.bed | awk -F'\t' '!seen[$1, $2, $3]++' > {{values.output_folder}}/{{values.genome_digest}}_annotations.bed
  - rm -f {{values.output_folder}}/{{values.genome_digest}}_enhancer.bed {{values.output_folder}}/{{values.genome_digest}}_promoter.bed {{values.output_folder}}/{{values.genome_digest}}_promoter_flanking.bed {{values.output_folder}}/{{values.genome_digest}}_5utr.bed {{values.output_folder}}/{{values.genome_digest}}_3utr.bed {{values.output_folder}}/{{values.genome_digest}}_exons.bed {{values.output_folder}}/{{values.genome_digest}}_introns.bed
  - gzip -f {{values.output_folder}}/{{values.genome_digest}}_annotations.bed
default_asset: "default"

# --- Additive non-runtime metadata (ignored by the builder) ---
tags:
- annotation
- genomic-features
- ensembl
outputs:
- pattern: '*_annotations.bed.gz'
  description: Combined genomic feature annotation file (gzipped)
test:
  commands:
  - test -f {output_dir}/{genome}_annotations.bed.gz
resources:
  memory: 4GB
  disk: 5GB
  time: 1h
metadata:
  author: nsheff
  created: '2026-05-20'
  license: BSD-2-Clause