From 792be78a8096eb111a090b573d64520a615bc760 Mon Sep 17 00:00:00 2001
From: Heng Li <lh3@me.com>
Date: Sun, 16 Nov 2014 00:46:02 -0500
Subject: [PATCH] replace run-gen-hs38* with run-gen-ref

---
 extras/run-gen-hs38a  |  9 ---------
 extras/run-gen-hs38d6 |  9 ---------
 extras/run-gen-ref    | 31 +++++++++++++++++++++++++++++++
 3 files changed, 31 insertions(+), 18 deletions(-)
 delete mode 100755 extras/run-gen-hs38a
 delete mode 100755 extras/run-gen-hs38d6
 create mode 100755 extras/run-gen-ref

diff --git a/extras/run-gen-hs38a b/extras/run-gen-hs38a
deleted file mode 100755
index cd119ad..0000000
--- a/extras/run-gen-hs38a
+++ /dev/null
@@ -1,9 +0,0 @@
-#!/bin/bash
-
-root=`dirname $0`
-
-wget -O- ftp://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Homo_sapiens/GRCh38/seqs_for_alignment_pipelines/GCA_000001405.15_GRCh38_full_analysis_set.fna.gz \
-  | gzip -dc > hs38a.fa
-
-[ ! -f hs38a.fa.alt ] && grep _alt $root/resource-GRCh38/hs38d6.fa.alt > hs38a.fa.alt
-[ ! -f hs38a.fa.bwt ] && echo -e "\nPlease run 'bwa index hs38a.fa'...\n"
diff --git a/extras/run-gen-hs38d6 b/extras/run-gen-hs38d6
deleted file mode 100755
index 86d6fa2..0000000
--- a/extras/run-gen-hs38d6
+++ /dev/null
@@ -1,9 +0,0 @@
-#!/bin/bash
-
-root=`dirname $0`
-
-(wget -O- ftp://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Homo_sapiens/GRCh38/seqs_for_alignment_pipelines/GCA_000001405.15_GRCh38_full_analysis_set.fna.gz \
-  | gzip -dc; cat $root/data/hs38d6-extra.fa) > hs38d6.fa
-
-[ ! -f hs38d6.fa.alt ] && cp $root/resource-GRCh38/hs38d6.fa.alt .
-[ ! -f hs38d6.fa.bwt ] && echo -e "\nPlease run 'bwa index hs38d6.fa'...\n"
diff --git a/extras/run-gen-ref b/extras/run-gen-ref
new file mode 100755
index 0000000..86317c3
--- /dev/null
+++ b/extras/run-gen-ref
@@ -0,0 +1,31 @@
+#!/bin/bash
+
+root=`dirname $0`
+
+url38="ftp://ftp.ncbi.nlm.nih.gov/genbank/genomes/Eukaryotes/vertebrates_mammals/Homo_sapiens/GRCh38/seqs_for_alignment_pipelines/GCA_000001405.15_GRCh38_full_analysis_set.fna.gz"
+url37d5="ftp://ftp.ncbi.nlm.nih.gov/1000genomes/ftp/technical/reference/phase2_reference_assembly_sequence/hs37d5.fa.gz"
+
+if [ $# -eq 0 ]; then
+	echo "Usage: $0 <hs38|hs38a|hs38d6|hs37|hs37d5>"
+	exit 1;
+fi
+
+if [ $1 == "hs38d6" ]; then
+	(wget -O- $url38 | gzip -dc; cat $root/resource-GRCh38/hs38d6-extra.fa) > $1.fa
+	[ ! -f $1.fa.alt ] && cp $root/resource-GRCh38/hs38d6.fa.alt $1.fa.alt
+elif [ $1 == "hs38a" ]; then
+	wget -O- $url38 | gzip -dc > $1.fa
+	[ ! -f $1.fa.alt ] && grep _alt $root/resource-GRCh38/hs38d6.fa.alt > $1.fa.alt
+elif [ $1 == "hs38" ]; then
+	# we don't use GCA_000001405.15_GRCh38_no_alt_analysis_set.fna.gz because it lacks EBV
+	wget -O- $url38 | gzip -dc | awk '/^>/{f=/_alt/?0:1}f' > $1.fa
+elif [ $1 == "hs37d5" ]; then
+	wget -O- $url37d5 | gzip -dc > $1.fa 2>/dev/null
+elif [ $1 == "hs37" ]; then
+	wget -O- $url37d5 | gzip -dc 2>/dev/null | awk '/^>/{f=/>hs37d5/?0:1}f' > $1.fa
+else
+	echo "ERROR: unknown genome build"
+fi
+
+[ ! -f $1.fa.bwt ] && echo -e "\nPlease run 'bwa index $1.fa'...\n"
+