From ea48d6fe09ce2760c04eee3a61670f591aa8d189 Mon Sep 17 00:00:00 2001
From: tcezard <tcezard@ebi.ac.uk>
Date: Mon, 15 Apr 2024 09:17:52 +0100
Subject: [PATCH] Separate RS from multiple lines

---
 .../gather_clustering_counts/bash/count_rs_for_all_files.sh    | 3 ++-
 1 file changed, 2 insertions(+), 1 deletion(-)

diff --git a/eva-accession-release-automation/gather_clustering_counts/bash/count_rs_for_all_files.sh b/eva-accession-release-automation/gather_clustering_counts/bash/count_rs_for_all_files.sh
index 4f8ee1b82..9ed94b2f8 100755
--- a/eva-accession-release-automation/gather_clustering_counts/bash/count_rs_for_all_files.sh
+++ b/eva-accession-release-automation/gather_clustering_counts/bash/count_rs_for_all_files.sh
@@ -19,7 +19,8 @@ do
     OUTPUT=tmp_${SC_NAME}_${ASSEMBLY}_${TYPE}.txt
     if [[ ${INPUT} == *.vcf.gz ]]
     then
-        zcat  "${INPUT}" | grep -v '^#' | awk -v annotation="${ASSEMBLY}-${SC_NAME}-${TYPE}" '{print $3" "annotation}' > ${OUTPUT}
+        # There are sometime multiple rs (separated by ;) in one line that needs to be split across multiple lines
+        zcat  "${INPUT}" | grep -v '^#' | awk -v annotation="${ASSEMBLY}-${SC_NAME}-${TYPE}" '{gsub(";","\n",$3); print $3" "annotation}' > ${OUTPUT}
     elif [[ ${INPUT} == *_unmapped_ids.txt.gz ]]
     then
         SC_NAME=$(basename $(dirname ${INPUT}));