6db87360227e298279ec6088c5e070f8c42e51b8
angie
  Tue Nov 14 17:11:45 2023 -0800
Masking some additional locations in BA.2 (& onward), BA.2.86 and XCK.

diff --git src/hg/utils/otto/sarscov2phylo/branchSpecificMask.yml src/hg/utils/otto/sarscov2phylo/branchSpecificMask.yml
index 0e3e5d7..1a51e1c 100644
--- src/hg/utils/otto/sarscov2phylo/branchSpecificMask.yml
+++ src/hg/utils/otto/sarscov2phylo/branchSpecificMask.yml
@@ -55,30 +55,33 @@
               [ 221, 225 ], [ 230, 233 ], [ 241, 243 ], [ 245, 246 ],
               # deletions
               [ 11288, 11296 ], [ 21633, 21641 ], [ 28361, 28371 ], [ 29734, 29759 ],
               # 28877 and 28878 together are highly homoplasic in all of B.1.1 (28881-28883).
               # They seem to be found very consistently in P.1*, but pop up in many places in Alpha
               # and Omicrons.  I haven't looked closely at the Alpha instances but they caused
               # some mini-Omicrons (https://github.com/cov-lineages/pango-designation/issues/988).
               # Possibly could also mask in B.1.1.7 and BA.1 but those are old news.
               [ 28877, 28878 ],
               # 3'UTR
               [ 29769, 29779 ], [ 29781, 29782 ] ]
     sites: [ # 5'UTR
              103, 110, 119, 121, 154, 162, 164, 214, 228, 239,
              # amplicon dropout (there are so many more; omitting to avoid recombinant trouble)
              22786, 22882, 23854,
+             # Recurrent multi-muts / misaligned insertions
+             # https://github.com/cov-lineages/pango-designation/issues/2327#issuecomment-1763481773
+             28245, 28251, 28254,
              # 3' UTR
              29760, 29762, 29764,
              29766,                   # only Luxembourg, made a mini-BA.2
              29767, 29784, 29786, 29793, 29800, 29803 ]
 
 BA.2.75:
     # Inherits BA.2 masking but has so many problems with false reversions I'm adding a ton here.
     representative: India/WB-INSACOG-1931503209307/2022
     reversions: [ # BA.2-level but not masked BA.2-wide to avoid messing up recombinants:
                   G670T, T2790C, T3037C, T4321C, G9424A, T9534C, T9866C, T10029C, T10198C,
                   G18163A, T19955C, G20055A, T21618C, G22200T, A22578G, T22674C, C22679T,
                   T22686C, G22688A, A22775G, T22813G, A22992G, A22995C, C23013A, G23055A,
                   T23063A, C23075T, G23403A, T23525C, G23599T, A23604C, T23948G, T24424A,
                   A24469T, T25000C, T26270C, G26577C, T27807C, T28271A, C29510A,
                   # BA.2.75-defining, very dropout-prone:
@@ -86,30 +89,40 @@
                   # BA.2.75-defining, bad but maybe not quite as bad:
                   T3796C, T3927C, T5183C, G12444A, A15451G, G22190A, A22331G, A22898G,
                   G22942T, C23013A ]
     # False muts in re-placed (from BA.5) recombinants:
     # XBD: A26275G
     # XBP: G22331A, G22577C, G22898A, A26275G
     # XBR: A22190G, G22331A, G22577C, G22898A, A26275G
     # XBS: A22190G, G22331A, G22577C, G22898A, A26275G
 
 BN.1.2.3:
     # Inherits from BA.2.75
     representative: England/QEUH-326228D4/2022
     sites: [ 337 # https://github.com/cov-lineages/pango-designation/issues/2016#issuecomment-1626159006
            ]
 
+BA.2.86:
+    # Inherits from BA.2
+    # @Over-There-Is requested 21610 - very messy indeed.
+    # https://github.com/sars-cov-2-variants/lineage-proposals/issues/606#issuecomment-1801095482
+    # @aviczhl2 pointed out some recurring reversions:
+    # https://github.com/sars-cov-2-variants/lineage-proposals/issues/1072
+    representative: OY747147.1
+    sites: [ 21610 ]
+    reversions: [ T21711C, C22032T, A22033C, G22034A, A23012G, G26610A ]
+
 BA.4:
     # BA.4 is placed on the BA.2 branch so it inherits all the BA.2 sites.
     representative: SouthAfrica/NICD-N41664/2022
     ranges: [ [ 686, 694 ], [ 21765, 21770 ] ]
 
 BA.5:
     # BA.5 is placed on the BA.2 branch so it inherits all the BA.2 sites.
     representative: England/PHEP-YYFJPAM/2022
     ranges: [ [ 21765, 21770 ] ]
     # Some of these should be reverted in recombinants, but we're pretty much past the point of
     # simultaneous Delta/Omicron and the noise from false reversions is so intolerable that we'll
     # just have to watch out for missing reversions when working with recombinants.
     # False muts in recombinants that were later re-placed in BA.2.75:
     # XBD: G12160A, T22917G, T23018G
     # XBP: G12160A,
@@ -157,15 +170,24 @@
     representative: England/LSPA-32578111/2022
     reversions: [ T22317G ]
 
 XBB.1.5:
     # Inherits from XBB.1
     # Don't believe reversions on 27915 once we're as far as XBB.1.*
     representative: England/BRBR-32671539/2022
     reversions: [ T27915G ]
 
 XBC:
     # Inherits nothing!  Should find out its deletions.
     # Cornelius Roemer requested to mask several reversions in
     # https://github.com/cov-lineages/pango-designation/issues/1100#issuecomment-1426502678
     representative: Philippines/PH-VUI-142736/2022
     reversions: [ G5584A, T13019C, T22329C, T25000C, C27718T, T28271A ]
+
+XCK:
+    # Inherits from XBB.1.5
+    # @FedeGueli pointed out that the usher tree had a very flaky 29729. TL;DR mafft is counfounding
+    # two nearby deletions, a new 29726 and the old 29734-29759, and making a false subst by getting
+    # the deletion boundaries wrong.  Mask 29729 here.
+    representative: USA/TX-CDC-QDX84451512/2023
+    representativeBacktrack: 2
+    sites: [ 29729 ]