gtf.yaml

# Ensembl GTF file distributed by Ensembl for hg38
# Cleans GTF file by converting chromosome names to standard names
# Uses https://github.com/dpryan79/ChromosomeMappings to remap the chromosome names
---
attributes:
  name: gtf
  version: 78
recipe:
  full:
    recipe_type: bash
    recipe_cmds:
      - |
        url=http://ftp.ensembl.org/pub/release-78/gtf/homo_sapiens/Homo_sapiens.GRCh38.78.gtf.gz
        mkdir -p rnaseq
        remap_url=http://raw.githubusercontent.com/dpryan79/ChromosomeMappings/master/GRCh38_ensembl2UCSC.txt
        wget --no-check-certificate -qO- $remap_url | awk '{if($1!=$2) print "s/^"$1"/"$2"/g"}' > remap.sed
        wget --no-check-certificate -qO- $url | gunzip | sed -f remap.sed | grep -v "*_*_alt" > rnaseq/hg38.gtf
        rm remap.sed
    recipe_outfiles:
    - rnaseq/hg38.gtf