b81e6071edc73b0d973b29689c9d2dd6c84b8e8d
angie
  Mon Oct 10 15:48:55 2022 -0700
Tighten -maxDiff since it is still a lot less stringent than the --max-path-length filtering applied after placement.

diff --git src/hg/utils/otto/sarscov2phylo/makeNewMaskedVcf.sh src/hg/utils/otto/sarscov2phylo/makeNewMaskedVcf.sh
index a5af428..3de9790 100755
--- src/hg/utils/otto/sarscov2phylo/makeNewMaskedVcf.sh
+++ src/hg/utils/otto/sarscov2phylo/makeNewMaskedVcf.sh
@@ -300,23 +300,23 @@
         >> $renaming
     set -o pipefail
 fi
 if [ -s newGisaid.filtered.fa ]; then
     zcat $gisaidDir/metadata_batch_$today.tsv.gz \
     | grep -Fwf <(fastaNames newGisaid.filtered.fa) \
     | tawk '$3 != "" {print $3 "\t" $1 "|" $3 "|" $5;}' \
         >> $renaming
 fi
 wc -l $renaming
 
 # Make masked VCF
 tawk '{ if ($1 ~ /^#/) { print; } else if ($7 == "mask") { $1 = "NC_045512v2"; print; } }' \
     $problematicSitesVcf > mask.vcf
 time cat <(twoBitToFa $ref2bit stdout) $alignedFa \
-| faToVcf -maxDiff=1000 \
+| faToVcf -maxDiff=200 \
     -excludeFile=exclude.ids \
     -verbose=2 stdin stdout \
 | vcfRenameAndPrune stdin $renaming stdout \
 | vcfFilter -excludeVcf=mask.vcf stdin \
 | gzip -c \
     > new.masked.vcf.gz