c6d93a375ba89f43a73c00a93dd329c3f6a3d4d2
angie
  Sat Dec 23 19:19:35 2023 -0800
Increase --max-path-length so we don't lose recent sequences with many legit mutations.

diff --git src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
index 8fdf8df..286b5a4 100755
--- src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
+++ src/hg/utils/otto/sarscov2phylo/updateCombinedTree.sh
@@ -46,31 +46,31 @@
     # $scriptDir/usherClusterRun.sh $today
     # Instead of the cluster, use Cheng's blazingly fast new usher-sampled:
     time $usher \
         -T 50 -A -e 5 \
         -i prevRenamed.pb \
         -v new.masked.vcf.gz \
         -o merged.pb \
         --optimization_radius 0 --batch_size_per_process 10 \
         > usher.addNew.log 2>usher-sampled.stderr
     # Branch-specific masking
     time $scriptDir/maskDelta.sh merged.pb merged.deltaMasked.pb
     # Prune samples with too many private mutations and internal branches that are too long.
     $matUtils extract -i merged.deltaMasked.pb \
         --max-parsimony 20 \
         --max-branch-length 60 \
-        --max-path-length 175 \
+        --max-path-length 225 \
         -O -o merged.deltaMasked.filtered.pb
     # matOptimize: used -r 8 -M2 until 2023-05-12, then switched to Cheng's recommended
     # -m 0.00000001 -M 4 (avoid identical-child-node problem in
     #  https://github.com/sars-cov-2-variants/lineage-proposals/issues/40)
     # The -M 4 allowed up to radius 32, and crazy things started happening all while I was
     # trying to get the tree cleaned up for pango-designation release 1.20 --> lineageTree.
     # After 2023-05-20, when I found that matOptimize had moved a big chunk of B.1 onto a
     # B.1.1.7 garbage branch, causing big trouble for lineageTree (23_05_18_updateLineageTreePb.txt).
     # After that I changed it back to -M 2 for my sanity. If the identical-child thing happens again,
     # then I'll probably just run matOptimize twice, with a small radius the second time.
     time $matOptimize \
         -T 80 -m 0.00000001 -M 2 -S move_log.filtered \
         -i merged.deltaMasked.filtered.pb \
         -o gisaidAndPublic.$today.masked.preTrim.pb \
         >& matOptimize.filtered.log