af90beb812e8f96e2283e62bd5082948ce455c58
angie
  Tue Oct 31 09:54:12 2023 -0700
Updating default column offsets to match more recent nextclade versions

diff --git src/hg/utils/otto/sarscov2phylo/findDropoutContam.pl src/hg/utils/otto/sarscov2phylo/findDropoutContam.pl
index 5a4bd4b..3786875 100755
--- src/hg/utils/otto/sarscov2phylo/findDropoutContam.pl
+++ src/hg/utils/otto/sarscov2phylo/findDropoutContam.pl
@@ -4,42 +4,42 @@
 # that have only a subset of Omicron mutations so we can keep them from screwing up
 # the base of the Omicron branch.
 
 # Some bad sequences are assigned 19A, 20A, 20B but have a suspicious number of Omicron muts.
 # Others are assigned Omicron (21K, 21L, 21M, 22*) but have a lot of reversions.
 
 use warnings;
 use strict;
 
 my $maxOmicronMuts = 5;
 my $maxReversions = 5;
 
 # Column offsets:
 #0       seqName
 #1       clade
-#18      privateNucMutations.reversionSubstitutions
-#19      privateNucMutations.labeledSubstitutions
-#30      nonACGTNs
+#22      privateNucMutations.reversionSubstitutions
+#23      privateNucMutations.labeledSubstitutions
+#34      nonACGTNs
 
 # Examples values for a seq assigned to 21J (Delta) but with a suspicious number of Omicron muts
 # (and Delta back-muts):
 # reversions example: T4181G,T7124C,T8986C,T9053G,T16466C,G21618C,C27638T,T27752C,T29402G
 # labeled example: T5386G|21K,G8393A|21K,C10449A|21K&21L&21M,A11537G|21K,T13195C|21K&21M,A17236G|21J,A18163G|21K&21L,C21762T|21K&21D&21M,C23525T|21K&20J&21L&21M,T23599G|21K&21L&21M,G23604A|20I&21K&21H&21L&21E&21M,G23948T|21K&21L,C24130A|21K&21M,C24503T|21K,A26530G|21K&21M,C26577G|21K&21L&21M,T27291C|21J,-28271T|21K&21G&21L&21M,C28311T|21K&21F&21G&21L&21M,T28881A|20I&21K&20B&20J&20F&20D&21G&21L&21E&21M
 
-my $reversionsIx = 18;
-my $labeledIx = 19;
-my $ambigIx = 30;
+my $reversionsIx = 22;
+my $labeledIx = 23;
+my $ambigIx = 34;
 
 sub cladeIsOmicron($) {
   my ($clade) = @_;
   return $clade =~ /^(21[KLM]|2[2-9]|recombinant)/;
 }
 
 sub reversionCount($$) {
   # Exclude ambiguous bases from reversions.  Aside from that:
   # Just return the number of mutations in the comma-sep list, no second-guessing, although
   # I've seen cases where a sequence is placed out at the end of a long branch and half of the
   # long branch muts are counted against it as reversions -- even though in the big tree, that
   # long branch is broken up many times and breaking it up would be usher's approach.  However,
   # in Nextclade's little tree, Omicron root is on a long branch, and in that case we do want
   # to count reversions against sequences that break up that particular long branch.
   # That's why I'm only looking at reversions (below) when the sequence is assigned to Omicron.