eeafd7b8b2ea5f34ee78bfe73b51ea7a1be5bce4
angie
  Thu Sep 2 17:51:46 2021 -0700
Add the same filters that we've been using on the combined tree because when GISAID samples are removed, some branches become very long.  (Russ & Yatish request)

diff --git src/hg/utils/otto/sarscov2phylo/extractPublicTree.sh src/hg/utils/otto/sarscov2phylo/extractPublicTree.sh
index 3ee55c2..5a8a750 100755
--- src/hg/utils/otto/sarscov2phylo/extractPublicTree.sh
+++ src/hg/utils/otto/sarscov2phylo/extractPublicTree.sh
@@ -18,30 +18,32 @@
 ottoDir=/hive/data/outside/otto/sarscov2phylo
 cncbDir=$ottoDir/cncb.latest
 
 scriptDir=$(dirname "${BASH_SOURCE[0]}")
 source $scriptDir/util.sh
 
 usherDir=~angie/github/usher
 matUtils=$usherDir/build/matUtils
 
 cd $ottoDir/$today
 
 # Extract public samples from tree
 grep -v EPI_ISL_ samples.$today > newPublicNames
 $matUtils extract -i gisaidAndPublic.$today.masked.pb \
     -s newPublicNames \
+    --max-parsimony 20 \
+    --max-branch-length 30 \
     -O -o public-$today.all.masked.pb
 
 # Add nextclade annotations to public protobuf
 if [ -s cladeToName ]; then
     # Use combined tree's clade assignments to annotate clades on public tree
     grep -v EPI_ISL cladeToName > cladeToPublicName
     time $matUtils annotate -T 50 \
         -l \
         -i public-$today.all.masked.pb \
         -c cladeToPublicName \
         -f 0.95 \
         -D details.nextclade.public \
         -o public-$today.all.masked.nextclade.pb \
         >& annotate.nextclade.public
 else