c9446936b026a803d97e7f5128404a327829fe2e
max
  Sun Jul 19 00:01:22 2026 -0700
ClinVar Mapped: add Pfam-domain mapping method

Maps ClinVar coding variants through Pfam protein domains as a second method
under the clinvarMapped container. Each Pfam family's profile HMM gives a shared
coordinate, so a variant inside a domain is projected to the residue in the same
HMM match-state column in every other MANE Select gene carrying that domain, then
mapped back to that gene's genomic codon. This links genes that share a domain
without being paralogs (e.g. the SCN sodium channels and the CACNA1 calcium
channels via Ion_trans). Each variant is shown at its 25 most conserved
equivalents. Uses Pfam-A 38.2. Adds two alpha subtracks (Pfam Domain Variants,
Pfam Domains), the hmmsearch/hmmalign build scripts, autoSql, makeDoc and
description page, refs #37883

diff --git src/hg/makeDb/trackDb/human/hg38/clinvarMapped.ra src/hg/makeDb/trackDb/human/hg38/clinvarMapped.ra
index 6526aa731aa..b08d8ebd216 100644
--- src/hg/makeDb/trackDb/human/hg38/clinvarMapped.ra
+++ src/hg/makeDb/trackDb/human/hg38/clinvarMapped.ra
@@ -1,23 +1,23 @@
 track clinvarMapped
 compositeTrack on
 shortLabel ClinVar Mapped
 longLabel ClinVar coding variants mapped to related loci by sequence homology
 group phenDis
 visibility hide
 type bigBed
-dataVersion Ensembl 116 paralogs, MANE Select 1.5, ClinVar as of build date
+dataVersion Ensembl 116 paralogs, Pfam-A 38.2, MANE Select 1.5, ClinVar as of build date
 html clinvarMapped
 
     track clinvarMappedParalog
     parent clinvarMapped on
     shortLabel Paralog Variants
     longLabel ClinVar coding variants mapped to the equivalent residue of paralogous genes
     type bigBed 12 +
     bigDataUrl /gbdb/hg38/clinvarMapped/clinvarMappedParalog.bb
     itemRgb on
     noScoreFilter on
     maxItems 100000
     urls vcvId="https://www.ncbi.nlm.nih.gov/clinvar/variation/$$/" sourceLocus="hgTracks?db=hg38&position=$$"
     mouseOver Mapped from ${sourceGene} ${sourceVariant} &mdash; ${clinSign} (${reviewStars} star)<br>Source residue ${srcRes}${sourceAaPos} in ${sourceGene}; aligned residue here is ${thisRes} (${residueMatch})<br>Ensembl paralog identity ${bioPercId}%, alignment identity ${alnPercId}%
     filterText.sourceGene *
     filterType.sourceGene wildcard
@@ -46,15 +46,61 @@
     filterLabel.alnPercId Alignment percent identity
 
     track clinvarMappedParalogAln
     parent clinvarMapped off
     shortLabel Paralog Alignments
     longLabel Paralog protein alignments used to map the variants
     type bigPsl
     bigDataUrl /gbdb/hg38/clinvarMapped/clinvarMappedParalogAln.bb
     visibility pack
     baseColorUseSequence lfExtra
     baseColorDefault diffBases
     showDiffBasesAllScales .
     showDiffBasesMaxZoom 10000.0
     indelDoubleInsert on
     indelQueryInsert on
+
+    track clinvarMappedPfam
+    parent clinvarMapped off
+    shortLabel Pfam Domain Variants
+    longLabel ClinVar coding variants mapped to the equivalent Pfam domain position in other genes
+    type bigBed 12 +
+    bigDataUrl /gbdb/hg38/clinvarMapped/clinvarMappedPfam.bb
+    itemRgb on
+    noScoreFilter on
+    maxItems 100000
+    urls vcvId="https://www.ncbi.nlm.nih.gov/clinvar/variation/$$/" sourceLocus="hgTracks?db=hg38&position=$$" pfamAcc="https://www.ebi.ac.uk/interpro/entry/pfam/$$/"
+    mouseOver Mapped from ${sourceGene} ${sourceVariant} &mdash; ${clinSign} (${reviewStars} star)<br>${pfamId} domain position ${domainColumn}: source residue ${srcRes}${sourceAaPos} in ${sourceGene}, residue here is ${thisRes} in ${thisGene} (${residueMatch})
+    filterText.sourceGene *
+    filterType.sourceGene wildcard
+    filterLabel.sourceGene Source gene (wildcards allowed, e.g. SCN5A or SCN*)
+    filterText.pfamId *
+    filterType.pfamId wildcard
+    filterLabel.pfamId Pfam domain (wildcards allowed)
+    filterValues.clinSignCode PG|pathogenic,LP|likely pathogenic,CF|conflicting,VUS|uncertain,RF|risk factor,OT|other,LB|likely benign,BN|benign
+    filterValuesDefault.clinSignCode PG,LP
+    filterType.clinSignCode multipleListOr
+    filterLabel.clinSignCode Clinical significance
+    filterValues.residueMatch identical,similar,different
+    filterType.residueMatch multipleListOr
+    filterLabel.residueMatch Residue conservation
+    filterValues.molConseq missense variant,nonsense,inframe deletion,inframe insertion,inframe indel,initiator codon variant,stop lost
+    filterType.molConseq multipleListOr
+    filterLabel.molConseq Source molecular consequence
+    filterByRange.reviewStars on
+    filter.reviewStars 0:4
+    filterLimits.reviewStars 0:4
+    filterLabel.reviewStars ClinVar review stars
+
+    track clinvarMappedPfamAln
+    parent clinvarMapped off
+    shortLabel Pfam Domains (MANE)
+    longLabel Pfam domain footprints on MANE Select genes, evidence for the Pfam mapping
+    type bigBed 12 +
+    bigDataUrl /gbdb/hg38/clinvarMapped/clinvarMappedPfamAln.bb
+    visibility pack
+    filterText.pfamId *
+    filterType.pfamId wildcard
+    filterLabel.pfamId Pfam domain (wildcards allowed)
+    filterText.gene *
+    filterType.gene wildcard
+    filterLabel.gene Gene (wildcards allowed)