78cdae7249c8609dcbc743e996ea7e5eec33d75a
max
  Mon Aug 17 08:15:39 2026 -0700
lrSv: fix off-by-one anchor base in deletion coordinates across converters, refs #38099

VCF/pangenome deletions carry a non-deleted anchor (padding) base at POS.
Several lrSv converters set chromStart = pos-1, which includes that anchor, so
each deletion was 1 bp too wide on the left and svLen was 1 too big. Callsets
handled this inconsistently, so the same deletion appeared at offset coordinates
and failed to merge in lrSvAll.

For deletions only (INS/INV/CPX unchanged), advance chromStart past the anchor
so the interval covers exactly the deleted bases (svLen == |SVLEN|). Verified
against the hg38 reference: the old left base is present in both REF and ALT
(i.e. retained by the sample), so it should not be inside the deletion.

Fixed 11 converters: lrSv1kLin1218VcfToBed, lrSv1kgOntVcfToBed,
lrSvGustafsonVcfToBed, lrSvGa4kSvVcfToBed, lrSvDecodeVcfToBed,
lrSvAou1kCsvToBed, lrSvColorsDbSvVcfToBed, lrSvCardBbToBed, lrSvAprVcfToBed,
lrSvCpc1VcfToBed, lrSvVcfToBed (generic, used by han945).

Left unchanged, verified already anchor-correct: hgsvc3 and hgsvc2 (0-based
source), hprc2v21 (Ro converter prefix-trims), noyvert/tommoJp (POS is the
first deleted base), chirmade101 (1-based-closed source).

Rebuilt all affected bigBeds (hg38 + hs1 where present) and the lrSvAll merge:
3,111,026 -> 2,963,093 rows as ~148k duplicate deletions now merge.

diff --git src/hg/makeDb/scripts/lrSv/lrSvDecodeVcfToBed.py src/hg/makeDb/scripts/lrSv/lrSvDecodeVcfToBed.py
index 8526967ed67..b6bbd96dda5 100644
--- src/hg/makeDb/scripts/lrSv/lrSvDecodeVcfToBed.py
+++ src/hg/makeDb/scripts/lrSv/lrSvDecodeVcfToBed.py
@@ -1,111 +1,116 @@
 #!/usr/bin/env python3
 """Convert the deCODE high-confidence long-read SV VCF to BED9+ for bigBed.
 
 Usage:
     lrSvDecodeVcfToBed.py input.vcf.gz output.bed
 
 The source VCF has no allele count, so the AC column is left empty (it used to
 carry a fake placeholder of 50). The source also lists some variants more than
 once (byte-identical records); those duplicate rows are collapsed here.
 """
 
 import gzip
 import os
 import sys
 
 sys.path.insert(0, os.path.dirname(os.path.abspath(__file__)))
 from lrSvCommon import svName, normalizeSvType, svColor
 
 
 def parseInfo(infoStr):
     d = {}
     for item in infoStr.split(";"):
         if "=" in item:
             k, v = item.split("=", 1)
             d[k] = v
         else:
             d[item] = True
     return d
 
 
 def main():
     if len(sys.argv) != 3:
         print(__doc__, file=sys.stderr)
         sys.exit(1)
 
     inFile, outFile = sys.argv[1], sys.argv[2]
     opener = gzip.open if inFile.endswith(".gz") else open
 
     seen = set()
     nIn = 0
     nDup = 0
     with opener(inFile, "rt") as fIn, open(outFile, "w") as fOut:
         for line in fIn:
             if line.startswith("#"):
                 continue
             nIn += 1
 
             fields = line.rstrip("\n").split("\t")
             chrom = fields[0]
             pos = int(fields[1])
             info = parseInfo(fields[7])
 
             svTypeRaw = info.get("SVTYPE", ".")
             svType = normalizeSvType(svTypeRaw)
             end = int(info.get("END", pos))
             svLenRaw = int(float(info.get("SVLEN", "0")))
             trrBegin = info.get("TRRBEGIN", ".")
             trrEnd = info.get("TRREND", ".")
 
             if trrBegin == ".":
                 trrBegin = ""
             if trrEnd == ".":
                 trrEnd = ""
 
             chromStart = pos - 1
             chromEnd = end
+            # POS is the non-deleted anchor base; drop it from the left of DEL
+            # intervals so svLen == |SVLEN| and coordinates match anchor-excluded
+            # callsets. INS/other keep the anchor-based position.
+            if svType == "DEL":
+                chromStart += 1
             if chromEnd <= chromStart:
                 chromEnd = chromStart + 1
 
             svLen = chromEnd - chromStart
             if svType in ("INS", "MEI"):
                 insLen = abs(svLenRaw)
             else:
                 insLen = 0  # DEL/INSDEL
 
             color = svColor(svType)
 
             featLen = insLen if svType in ("INS", "MEI") else svLen
             # deCODE publishes no allele count, so omit AC from the name.
             name = svName(svType, featLen)
 
             row = [
                 chrom,
                 str(chromStart),
                 str(chromEnd),
                 name,
                 "0",
                 ".",
                 str(chromStart),
                 str(chromEnd),
                 color,
                 svType,
                 str(svLen),
                 str(insLen),
                 "",          # AC: deCODE has none (was a fake placeholder of 50)
                 trrBegin,
                 trrEnd,
             ]
             line_out = "\t".join(row)
             if line_out in seen:
                 nDup += 1
                 continue
             seen.add(line_out)
             fOut.write(line_out + "\n")
 
     print(f"deCODE: {nIn:,} input records, {nDup:,} duplicate rows dropped, "
           f"{nIn - nDup:,} written", file=sys.stderr)
 
 
 if __name__ == "__main__":
     main()