2ba814d9a2b0df7bdc2cce0bb3e52464288a61bc
angie
  Tue Jul 13 09:44:37 2021 -0700
no need to run matUtils extract -u twice.

diff --git src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
index dcca06e..bc8966b 100755
--- src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
+++ src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
@@ -466,32 +466,31 @@
 
 # EPI_ISL_ ID to public sequence name mapping, so if users upload EPI_ISL IDs for which we have
 # public names & IDs, we can match them.
 cut -f 1,3 $epiToPublic > epiToPublic.latest
 
 # Update links to latest public+GISAID protobuf and metadata in hgwdev cgi-bin directories
 for dir in /usr/local/apache/cgi-bin{-angie,-beta,}/hgPhyloPlaceData/wuhCor1; do
     ln -sf `pwd`/gisaidAndPublic.$today.masked.pb $dir/public.plusGisaid.latest.masked.pb
     ln -sf `pwd`/gisaidAndPublic.$today.metadata.tsv.gz \
         $dir/public.plusGisaid.latest.metadata.tsv.gz
     ln -sf `pwd`/hgPhyloPlace.plusGisaid.description.txt $dir/public.plusGisaid.latest.version.txt
     ln -sf `pwd`/epiToPublic.latest $dir/
 done
 
 # Extract public samples from tree
-$matUtils extract -i gisaidAndPublic.$today.masked.pb -u newNames
-grep -v EPI_ISL_ newNames > newPublicNames
+grep -v EPI_ISL_ samples.$today > newPublicNames
 $matUtils extract -i gisaidAndPublic.$today.masked.pb \
     -s newPublicNames \
     -O -o public-$today.all.masked.pb
 # Extract Newick and VCF from public-only tree
 $matUtils extract -i public-$today.all.masked.pb \
     -t public-$today.all.nwk \
     -v public-$today.all.masked.vcf
 gzip -f public-$today.all.masked.vcf
 zcat gisaidAndPublic.$today.metadata.tsv.gz \
 | grep -v EPI_ISL_ \
 | gzip -c \
     > public-$today.metadata.tsv.gz
 
 grep -v EPI_ISL_ cladeToName > cladeToPublicName
 grep -v EPI_ISL_ lineageToName > lineageToPublicName