b9b84244ae9fbfa6d81c92028634be26e8a5898f
angie
  Tue Jan 5 00:50:55 2021 -0800
Prevent redundant seqToEpi lines from nextmeta cobbled together from old nextmeta plus manually downloaded sequences.

diff --git src/hg/utils/otto/sarscov2phylo/updateIdMapping.sh src/hg/utils/otto/sarscov2phylo/updateIdMapping.sh
index be70d07..00ffc31 100755
--- src/hg/utils/otto/sarscov2phylo/updateIdMapping.sh
+++ src/hg/utils/otto/sarscov2phylo/updateIdMapping.sh
@@ -34,31 +34,31 @@
 mkdir -p $cogUkDir
 cd $cogUkDir
 $scriptDir/getCogUk.sh >& getCogUk.log
 
 # Last time I checked, CNCB had not updated since September, just keep using what we have
 cncbDir=$ottoDir/cncb.latest
 
 # Set up input files for Chris's scripts to map GISAID <--> public sequences
 cd $mapScriptDir
 mkdir input/$today
 cd input/$today
 ln -sf $cncbDir/cncb.nonGenBank.fasta .
 ln -sf $ncbiDir/genbank.fa.xz .
 ln -sf $cogUkDir/cog_all.fasta.xz .
 ln -sf $nextfasta .
-xcat $nextmeta | tail -n+2 | cut -f1,3 > seqToEpi
+xcat $nextmeta | tail -n+2 | cut -f1,3 | uniq > seqToEpi
 
 cd $mapScriptDir
 ./build.sh -t $today
 
 cd $installDir
 
 gbToDate=$ncbiDir/gbToDate
 cogUkToDate=$cogUkDir/cogUkToDate
 cncbToDate=$cncbDir/cncbToDate
 
 join -t$'\t' -a 1 -1 2 -o 1.1,1.2,1.3,2.2 \
     <(sort -k2,2 ~/chris_ncov/epiToPublicIdName.$today.txt) \
     <(sort $gbToDate $cncbToDate $cogUkToDate) \
 | sort \
     > epiToPublicAndDate.$today