Fixes download_reference

Illumina · Apr 4, 2020 · 2a369c0 · 2a369c0
1 parent fc4f38c
commit 2a369c0
Showing 1 changed file with 15 additions and 7 deletions.
diff --git a/scripts/download_reference.sh b/scripts/download_reference.sh
@@ -18,10 +18,13 @@ fi
 output_dir=`dirname ${output_file}`
 genome_build=${2:-"37"}
 
+# Mitochondrial refseq moved to a different folder, hence the "mt_remote"
 if [ ${genome_build} = "37" ]; then
-	remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/Homo_sapiens/ARCHIVE/BUILD.37.3/Assembled_chromosomes/seq"
+	remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Homo_sapiens/all_assembly_versions/GCF_000001405.25_GRCh37.p13/GCF_000001405.25_GRCh37.p13_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/"
+    mt_remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Homo_sapiens/all_assembly_versions/GCF_000001405.25_GRCh37.p13/GCF_000001405.25_GRCh37.p13_assembly_structure/non-nuclear/assembled_chromosomes/FASTA/"
 elif [ ${genome_build} = "38" ]; then
-    remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/Homo_sapiens/ARCHIVE/ANNOTATION_RELEASE.109/Assembled_chromosomes/seq/"
+    remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Homo_sapiens/all_assembly_versions/GCF_000001405.39_GRCh38.p13/GCF_000001405.39_GRCh38.p13_assembly_structure/Primary_Assembly/assembled_chromosomes/FASTA/"
+    mt_remote="ftp://ftp.ncbi.nlm.nih.gov/genomes/refseq/vertebrate_mammalian/Homo_sapiens/all_assembly_versions/GCF_000001405.39_GRCh38.p13/GCF_000001405.39_GRCh38.p13_assembly_structure/non-nuclear/assembled_chromosomes/FASTA/"
 else
 	echo "Error: Unsupported genome build ${genome_build}, valid values are 37,38"
 	exit 1
@@ -30,16 +33,21 @@ fi
 temp_dir=`mktemp -d 2>/dev/null || mktemp -d -t ${output_dir}`
 pushd ${temp_dir}
 
-for chrom in `seq 1 22` X Y MT
+for chrom in `seq 1 22` X Y
 do
-    wget ${remote}/*_ref_*chr${chrom}.fa.gz
+    wget ${remote}/chr${chrom}.fna.gz
 done
+wget ${mt_remote}/chrMT.fna.gz
 
-for chrom in `seq 1 22` X Y MT
+build_fa(){
+    echo ">${1}" >> ${2}
+    gunzip -c chr${1}.fna.gz | grep -v ">" >> "${2}"
+}
+for chrom in `seq 1 22` X Y
 do
-    echo ">${chrom}" >> ${output_file}
-    gunzip -c *_ref_*chr${chrom}.fa.gz | grep -v ">" >> "${output_file}"
+    build_fa $chrom $output_file
 done
+build_fa "MT" $output_file
 
 if hash samtools 2>/dev/null; then
     samtools faidx ${output_file}