ed7e20316990e2e6e52e0baf73087222eb571b67
angie
  Fri Jan 20 12:58:02 2023 -0800
Mask reversion G26530A in BA.1.1 because it made a mess of the branch leading to BA.1.1.1.

diff --git src/hg/utils/otto/sarscov2phylo/maskDelta.sh src/hg/utils/otto/sarscov2phylo/maskDelta.sh
index 056d44b..0a66b35 100755
--- src/hg/utils/otto/sarscov2phylo/maskDelta.sh
+++ src/hg/utils/otto/sarscov2phylo/maskDelta.sh
@@ -195,20 +195,36 @@
 done >> $maskFile
 
 # BA.5 reversions causing big yuck branches
 for backMut in G670T T2790C T3037C A4184G T4321C G9424A T9534C T10198C A12160G T15714C T17410C \
     G18163A T19955C G20055A T21618C G22200T A22578G T22674C C22679T T22686C G22688A A22775G T22813G \
     G22882T G22917T A22992G A22995C C23013A G23018T G23055A T23063A C23075T G23403A T23525C G23599T \
     A23604C T25000C A26529G G26577C A26709G T27807C T27889C T28271A C29510A ; do
     echo -e "$backMut\t$BA5Node"
 done >> $maskFile
 # The path to XAV should have G12160A and C27889T, but those cause too much trouble in BA.5
 # so it will have to be a little incorrect.
 # Likewise, XAN path should have G12160A but oh well.
 
 set -x
 
+# BA.1.1 has a ton of false reversions of A26530G that cause many flip-flops on the branch
+# leading to BA.1.1.1, as well as a bunch of sequences with noise at 76/77.  As of Jan. 2023
+# I'm not worried about messing up potential recombinants and want a cleaner tree for assignment.
+# Although it does add a false A26530G to the path for XD, oh well.
+BA11Node=$(grep England/ALDP-2BEB0A0/2021 $samplePaths \
+         | awk '{print $NF;}' | sed -re 's/:.*//;')
+set +x
+for backMut in G26530A ; do
+    echo -e "$backMut\t$BA11Node"
+done >> $maskFile
+for ((i=76;  $i <= 77;  i++)); do
+    echo -e "N${i}N\t$BA11Node"
+done >> $maskFile
+
+set -x
+
 time $matUtils mask -i $treeInPb \
     -m $maskFile \
     -o $treeOutPb
 
 rm $samplePaths