614ca430db1c23f496ccc5655d27b469e1c53443
lrnassar
  Fri Jul 31 16:38:30 2026 -0700
Apply Cardiomyopathy VCEP feedback: BP7 splice-only, drop PS4 tracks, evidence-first Variant Evidence Summary. refs #37446

Per the CM VCEP's answers to the interpretation questions:
- BP7: remove the conservation (phyloP) requirement and set the SpliceAI cutoff
to < 0.1, per Walker 2023 (PMID 37352859).
- PS4: remove the Walsh 2017 gene-level OR and Atlas EF display tracks (the VCEP
computes PS4 from current cohorts). Scripts and data are retained in the tree;
the tracks are dropped from the hub trackDb.
- Rename the former provisional/computable-codes track to "Variant Evidence
Summary" and switch its mouseover to an evidence-first layout: each item leads
with the data and is tagged with the ACMG criterion it would support; no overall
classification is computed.
- Confirmed and reflected in the docs: EvRepo is the PS1/PM5 reference (Q1); the
PM4 last-exon / within-50-nt rule (Q2); CardioBoost stays informational (Q6);
the PM1 codon ranges are HCM-only (Q8); no functional-score dataset is currently
VCEP-approved (Q4).

diff --git src/hg/makeDb/doc/Cardiomyopathy.txt src/hg/makeDb/doc/Cardiomyopathy.txt
index d55bf901d20..2fdcad99aa9 100644
--- src/hg/makeDb/doc/Cardiomyopathy.txt
+++ src/hg/makeDb/doc/Cardiomyopathy.txt
@@ -1,504 +1,511 @@
 ##############################################################################
 # Cardiomyopathy VCEP Track Hub — build documentation
 #
 # Redmine: #37446
 # VCEP:    https://clinicalgenome.org/affiliation/50002/
 # CSpec:   https://cspec.genome.network/cspec/ui/svi/affiliation/50002
 #
 # 8 gene/disease specifications, all dated 2024-04-22:
 #   GN002  MYH7    v2.0.0  HCM + DCM (AD)
 #   GN095  MYBPC3  v1.0.0  HCM (AD) — only PVS1-applicable gene
 #   GN098  TNNI3   v1.0.0  HCM (AD)
 #   GN099  TNNT2   v1.0.0  HCM + DCM (AD)
 #   GN100  TPM1    v1.0.0  HCM (AD)
 #   GN101  ACTC1   v1.0.0  HCM (AD)
 #   GN102  MYL2    v1.0.0  HCM (AD)
 #   GN103  MYL3    v1.0.0  HCM (AD)
 #
 # ARVC genes are out of scope (separate ClinGen panel, affiliation 40003).
 #
 # Expert contacts:
 #   Lucas Bronicki, PhD, FACMG     — Cardiomyopathy VCEP Chair
 #   Haley Garrett, MPH             — Coordinator (Haley_Garrett@med.unc.edu)
 #   ClinVar submitter ID:          506161 ("ClinGen Cardiomyopathy Variant
 #                                  Curation Expert Panel")
 #
 # Calibration sources:
 #   Walsh 2017 (PMID 27532257; Genet Med; DOI 10.1038/gim.2016.90)
 #                                   — PS4 case-control reference (Tables S5A/S5B;
-#                                     60,706 ExAC samples; 7,855 cases)
+#                                     60,706 ExAC samples; 7,855 cases). NOTE: the
+#                                     Walsh-2017 PS4 track was RETIRED from the hub
+#                                     2026-07 per the VCEP (Q5).
 #   Walsh 2019 (PMID 30696458; Genome Medicine; DOI 10.1186/s13073-019-0616-z)
 #                                   — PM1 hotspot codon ranges (Table S4),
 #                                     per-gene NonTrunc ExAC denominators
 #                                     (Table S1), and 155 pre-EvRepo per-variant
 #                                     curations (Table S6)
 #
 # MANE Select transcripts (verified against /gbdb/hg38/mane/mane.bb):
 #   MYH7    NM_000257.4    NP_000248.2    chr14 (-)
 #   MYBPC3  NM_000256.3    NP_000247.2    chr11 (-)
 #   TNNT2   NM_001276345.2 NP_001263274.1 chr1  (-)
 #   TNNI3   NM_000363.5    NP_000354.4    chr19 (-)
 #   TPM1    NM_001018005.2 NP_001018005.1 chr15 (+)  ← only plus-strand gene
 #   ACTC1   NM_005159.5    NP_005150.1    chr15 (-)
 #   MYL2    NM_000432.4    NP_000423.2    chr12 (-)
 #   MYL3    NM_000258.3    NP_000249.1    chr3  (-)
 #
 # NOTE on transcripts: the hgVai annotation layer (B.6) and most tracks use the
 # MANE Select transcript above. Two deliberate exceptions:
 #   - TNNT2 PM1 codon range (B.1) and the Walsh-2019 TNNT2 curations (B.7c) use
 #     the classic NM_001001430.2 where the source data is keyed to it; TNNT2
 #     hg19 coordinates for those are derived by liftOver from hg38 (no hg19
 #     transcript alignment). The MANE PM1 range 89-189 is used for the PM1 track.
 ##############################################################################
 
 
 ##############################################################################
 # Layout
 ##############################################################################
 #
 # Working directory:                 /hive/users/lrnassar/claude/RM37446/
 #     hub.txt, genomes.txt
 #     hg38/trackDb.txt, hg19/trackDb.txt
 #     cardiomyopathy.html               (shared description page)
 #     cmp_downloads/                    (source data — checked at A.0 + cached)
 #     scripts_local_copy/               (insurance copy of all 12 build scripts;
 #                                        live copy is at ~/kent/...)
 #     cmpVCEPClinDomains/               (PM1 hotspot regions)
 #     cmpVCEPPVS1/                      (MYBPC3-only PVS1 caveats)
 #     cmpVCEPAFfrequencies/             (gnomAD v4.1 BA1/BS1/PM2_supporting)
 #     cmpVCEPAnnotate/                  (hgVai consequence + HGVSp annotation TSV)
 #     cmpVCEPRevel/                     (REVEL PP3/BP4)
 #     cmpVCEPCardioBoost/               (CardioBoost missense predictor — off)
 #     cmpVCEPEvRepo/                    (VCEP Curated Variants — EvRepo, with codes)
 #     cmpVCEPClinVar506161/             (VCEP ClinVar submissions, no codes)
 #     cmpVCEPWalsh2019/                 (Walsh 2019 Pre-EvRepo Table S6 curations)
-#     cmpVCEPWalshOR/                   (Walsh 2017 gene-level OR — PS4)
+#     cmpVCEPWalshOR/                   (Walsh 2017 gene-level OR — PS4; RETIRED from hub, Q5)
 #     cmpVCEPAtlasEF/                   (Atlas of Cardiac Genetic Variation
-#                                        per-variant + UCSC-computed OR — PS4)
-#     cmpVCEPProvisionalClass/          (Computable ACMG Criteria Summary; per-variant codes)
+#                                        per-variant + UCSC OR — PS4; RETIRED from hub, Q5)
+#     cmpVCEPProvisionalClass/          (Variant Evidence Summary; per-variant evidence + criteria)
 #
 # Build scripts:                     ~/kent/src/hg/makeDb/scripts/cardiomyopathyVCEP/
 #     cmpVCEPClinDomains.py             (B.1 — PM1 hotspot regions; also provides
 #                                        parse_mane_record, cds_exons,
 #                                        aa_to_genomic_segments helpers imported
 #                                        by B.2, B.6, B.9, B.11)
 #     cmpVCEPPVS1.py                    (B.2)
 #     cmpVCEPAFfrequencies.py           (B.3; fetch_gene_variants() defines the
 #                                        shared variant universe reused by B.6/B.11)
 #     cmpVCEPAnnotate.py                (B.6 — hgVai consequence/HGVSp annotation)
 #     cmpVCEPRevel.py                   (B.4)
 #     cmpVCEPCardioBoost.py             (B.5)
 #     cmpVCEPEvRepo.py                  (B.7)
 #     cmpVCEPWalsh2019.py               (B.7c)
 #     cmpVCEPClinVar506161.py           (B.8)
 #     cmpVCEPWalshOR.py                 (B.9)
 #     cmpVCEPAtlasEF.py                 (B.10)
 #     cmpVCEPProvisionalClass.py        (B.11)
 #
 # Otto cron staging dir:             /hive/data/outside/otto/cardiomyopathyVCEP/
 #                                    (NOT YET WRITTEN — Phase E TODO; mirror TP53)
 
 
 ##############################################################################
 # Phase A: Source data
 ##############################################################################
 #
 # Working set lives in /hive/users/lrnassar/claude/RM37446/cmp_downloads/.
 # Most files are downloaded once and re-used; EvRepo + ClinVar 506161 are the
 # only sources intended for the weekly otto refresh (Phase E). Atlas is a
 # one-time snapshot (ExAC-based; does not change weekly).
 #
 # A.0  MANE Select transcript verification + Walsh-paper roles per gene
 #      (FIRST ACTION — propagates through all downstream phases)
 #
 #      Extract MANE bigGenePred for our 8 genes:
 #        bigBedToBed /gbdb/hg38/mane/mane.bb stdout \
 #          | awk -F'\t' -v OFS='\t' '
 #              BEGIN{print "geneSym\tchrom\tstart\tend\tstrand\trefSeqAcc\trefSeqProt\tensProtAcc"}
 #              $19 ~ /^(MYH7|MYBPC3|TNNT2|TNNI3|TPM1|ACTC1|MYL2|MYL3)$/ {
 #                print $19, $1, $2, $3, $6, $22, $24, $21
 #              }' | sort > cmp_downloads/mane_8genes.tsv
 #
 #      Walsh paper roles confirmed by parsing each CSpec HTML:
 #        PS4 source       = Walsh 2017 universally (8/8 genes)
 #        PM1 calibration  = Walsh 2019 (only MYH7, MYBPC3, TNNT2, TNNI3 — the
 #                           4 genes with a defined PM1 region)
 #
 #
 # A.1  CSpec HTML pages (8 docs)
 #
 #      mkdir -p cmp_downloads/cspec
 #      for gn in GN002 GN095 GN098 GN099 GN100 GN101 GN102 GN103; do
 #        curl -fsSL "https://cspec.genome.network/cspec/ui/svi/doc/${gn}" \
 #          -o cmp_downloads/cspec/${gn}.html
 #      done
 #
 #      Thresholds transcribed from the CSpec HTML (every value taken directly
 #      from the spec; none invented):
 #        BA1 universal:                >= 0.001    (FAF95 popmax)
 #        BS1 universal (7 of 8):       >= 0.0001
 #        BS1 MYBPC3 outlier:           >= 0.0002    (per GN095)
 #        PM2_supporting universal:     <= 4e-05
 #        PP3 REVEL universal:          >= 0.70
 #        BP4 REVEL universal:          <= 0.40
 #        PS4 OR (CI-lower) strength:   Strong >=20, Moderate >=10, Supporting >=5
 #        PM1 codon ranges:
 #          MYH7    167-931      (Walsh 2019 Table S4)
 #          MYBPC3  485-502 + 1248-1266
 #          TNNT2   89-189       (MANE NM_001276345.2; classic NM_001001430.2 = 79-179)
 #          TNNI3   141-209
 #          (TPM1, ACTC1, MYL2, MYL3 — no PM1 region defined)
 #        MYBPC3 PVS1 caveats:
 #          NMD-escape boundary: codon 1254+
 #          Micro-exons:         10, 11, 14
 #          In-frame exons:      2, 3, 4, 8, 9, 10, 11, 14, 20, 22, 24, 25, 26, 27
 #
 #      CSpec points where the spec is silent / leaves a curator choice (provisional
 #      decisions made for the build; posed to the VCEP for confirmation):
 #        - PS1/PM5 reference DB of "established pathogenic" (spec says "per
 #          Richards 2015", names none) — build uses VCEP EvRepo P/LP, leave-one-out.
 #        - PM4 for MYH7 (PVS1 N/A): spec redirects PM4 to non-NMD truncating at
 #          Moderate/Supporting, no boundary given — build uses last exon OR within
 #          50 nt of the final exon-exon junction.
-#        - BP7 "not highly conserved": no metric/cutoff in spec — build uses
-#          phyloP470way <= 0 together with SpliceAI < 0.20.
+#        - BP7 "not highly conserved": the VCEP removed the conservation requirement
+#          and set BP7 at SpliceAI < 0.1 (Walker 2023, PMID 37352859).
 #
 #
 # A.2  ClinGen EvRepo classifications (REST JSON)
 #
 #      curl -fsSL 'https://erepo.genome.network/evrepo/api/classifications?expertpanel=Cardiomyopathy+VCEP&format=json' \
 #        -o cmp_downloads/erepo/cardiomyopathyVCEP_classifications.json
 #
 #      State at 2026-06-29 refresh: 25 variants (all MYH7). One change vs the
 #      April pull — CAR:CA016422 moved Likely Pathogenic -> Uncertain Significance.
 #      April JSON retained as *.april2026.json.bak.
 #
 #
 # A.3  ClinVar submitter 506161 (filter from weekly VCV release)
 #
 #      Precise column-10 match against the submission_summary (loose grep over-
 #      counts because the submitter name also appears in other submitters' text):
 #        zcat /hive/data/outside/otto/clinvar/downloads/$LATEST/submission_summary.txt.gz \
 #          | awk -F'\t' '$10 == "ClinGen Cardiomyopathy Variant Curation Expert Panel"' \
 #          > cmp_downloads/clinvar/cardiomyopathyVCEP_submissions.tsv
 #
 #      State at ClinVar release 2026-05-30: 199 records, 100% "reviewed by
 #      expert panel". Classifications: 33 P + 32 LP + 75 VUS + 9 LB + 50 B = 199.
 #
 #
 # A.4  gnomAD v4.1 exomes (already on hgwdev; no download)
 #      /hive/data/outside/gnomAD.4/v4.1/exomes/gnomad.exomes.v4.1.sites.chr{N}.vcf.bgz
 #      Field used: fafmax_faf95_max (max FAF95 across genetic ancestry groups),
 #      queried per-region via tabix.
 #      Public mirror: https://hgdownload.soe.ucsc.edu/gbdb/hg38/gnomAD/v4.1/exomes/exomes.bb
 #
 # A.5  REVEL (already on hgwdev; no download)
 #      /gbdb/hg38/revel/{a,c,g,t}.bw  (per-alt-nucleotide bigwigs)
 #      Public mirror: https://hgdownload.soe.ucsc.edu/gbdb/hg38/revel/
 #
 # A.6  SpliceAI (already on hgwdev; no download)
 #      /gbdb/hg38/bbi/spliceAi.bb  (bed9+4; AIscore in col 9, name="ref>alt")
-#      Used by the B.11 SpliceAI safety net and BP7.
+#      Used by the B.11 splice flag and BP7 (SpliceAI < 0.1).
 #
 # A.7  CardioBoost precomputed predictions
 #      cmp_downloads/cardioboost/cm_prediction.RData  (precomputed table,
 #        ~65k rows; ~31k in our 8 genes — NOT just model objects)
 #      Source: https://github.com/ImperialCardioGenetics/CardioBoost_manuscript
 #      Loaded via /usr/bin/Rscript (the team Rscript is broken on hgwdev —
 #      missing libgfortran.so.3). Coordinates are GRCh37 with numeric chrom
 #      names; B.5 adds the chr prefix and liftOvers to hg38.
 #
 # A.8  Walsh PS4 + PM1 calibration supplements
 #      mkdir -p cmp_downloads/walsh ; cd cmp_downloads/walsh
 #      # Walsh 2017 (PS4 case-control) — Springer direct (avoids PMC PoW challenge)
 #      curl -fsSL -o walsh2017_supplement.zip \
 #        "https://static-content.springer.com/esm/art%3A10.1038%2Fgim.2016.90/MediaObjects/41436_2017_BFgim201690_MOESM9_ESM.zip"
 #      unzip -o walsh2017_supplement.zip   # -> Supplementary_Tables_resubmit.xlsx
 #      # Walsh 2019 (PM1 calibration + NonTrunc denoms + Table S6 curations)
 #      curl -fsSL -o walsh2019_supplement.xlsx \
 #        "https://static-content.springer.com/esm/art%3A10.1186%2Fs13073-019-0616-z/MediaObjects/13073_2019_616_MOESM1_ESM.xlsx"
 #      Walsh 2017 Tables S5A (HCM) / S5B (DCM): gene x disease x variant-class
 #        case-control OR + 95% CI (the PS4 source, B.9).
 #      Walsh 2019 Table S4: PM1 codon ranges. Table S1: per-gene NonTrunc ExAC
 #        denominators (B.10). Table S6: 155 per-variant ACMG curations (B.7c).
 #
 # A.9  Atlas of Cardiac Genetic Variation — one-time scrape with on-disk cache
 #      python3 cmp_downloads/atlas/scrape_atlas.py
 #      Two-pass (listing pages + per-variant pages for case_count >= 3), cached
 #      under cmp_downloads/atlas/raw/ + variants/, rate-limited 1.5 req/sec.
 #      The Atlas is a static companion to Walsh 2017 (ExAC controls, ~2016); it
 #      is NOT refreshed by otto. To refresh manually, delete the cache and re-run.
 
 
 ##############################################################################
 # Phase B: Build tracks
 ##############################################################################
 #
 # All build scripts at ~/kent/src/hg/makeDb/scripts/cardiomyopathyVCEP/
 # Each accepts: --db hg38 --db hg19 --output-dir <path>  (B.6 takes only
 # --output-dir). Each emits .as / .bed / .bb for both assemblies and verifies
 # cross-assembly parity at the end.
 #
 # Build order matters: B.3 defines the variant universe that B.6 annotates;
 # B.11 (Computable codes) consumes B.6 (annotation), B.3 (AF), B.4 (REVEL), and the
 # B.7 EvRepo P/LP set (PS1/PM5 reference). Suggested order:
 #
 #   for s in cmpVCEPClinDomains cmpVCEPPVS1 cmpVCEPAFfrequencies cmpVCEPAnnotate \
 #            cmpVCEPRevel cmpVCEPCardioBoost cmpVCEPEvRepo cmpVCEPWalsh2019 \
 #            cmpVCEPClinVar506161 cmpVCEPWalshOR cmpVCEPAtlasEF \
 #            cmpVCEPProvisionalClass; do
 #     python3 ~/kent/src/hg/makeDb/scripts/cardiomyopathyVCEP/${s}.py \
 #       --db hg38 --db hg19 \
 #       --output-dir /hive/users/lrnassar/claude/RM37446 || break
 #   done
 #   # (cmpVCEPAnnotate.py takes only --output-dir.)
 #
 #
 # B.1  cmpVCEPClinDomains.py — PM1 Hotspot Regions
 #      Output: 28 features (4 genes x CDS-exon-spanning AA range; MYH7, MYBPC3,
 #      TNNT2, TNNI3 only). Codon ranges per A.1. Color 230,3,131 (magenta-rose,
 #      the InSiGHT/TP53 clinical-domains convention). Built-in unit tests check
 #      codon-to-genomic conversion across all 8 genes.
 #
 # B.2  cmpVCEPPVS1.py — MYBPC3 PVS1 Evidence (MYBPC3 only)
 #      Output: 16 features (2 NMD-escape genomic segments + 14 in-frame exon
 #      caveats; 3 also tagged micro-exon in the mouseover).
 #
 # B.3  cmpVCEPAFfrequencies.py — gnomAD v4.1 BA1/BS1/PM2_supporting
 #      Variant universe: all gnomAD v4.1 PASS variants in the 8 gene CDS regions
 #      +/- 20 nt splice padding. fetch_gene_variants() is the canonical universe,
 #      reused by B.6 and B.11.
 #      Output: 10,974 features. Applied code: PM2_supporting 9,893, no-code 504,
 #      BS1 367, BA1 210. MYBPC3 BS1 outlier (0.0002) handled per gene.
 #
 # B.6  cmpVCEPAnnotate.py — hgVai consequence + HGVSp annotation layer
 #      Reuses B.3 fetch_gene_variants() for an identical universe, then runs
 #      hgVai per gene via vai.pl on the single MANE Select transcript:
 #        vai.pl --variantLimit=200000000 --hgVai=/usr/local/apache/cgi-bin/hgVai \
 #               --position=<chrom>:<txStart>-<txEnd> --geneTrack=ncbiRefSeqSelect \
 #               --hgvsG=off --hgvsCN=off --hgvsP=on hg38 <variants.vcf.gz>
 #      Each input VCF row carries ID=chrom:pos:ref:alt so indels join exactly on
 #      the way back (without the ID, VEP reformats indel names and ~659 fail to
 #      join). Output cmpVCEPAnnotate/cmpVCEPAnnotations.hg38.tsv: 10,974 rows,
 #      0 unmapped. Columns: chrom,pos,ref,alt,gene,soTerms,proteinPos,aaRef,aaAlt,
 #      codonChange,exonNum,exonTotal,cdnaPos,hgvsp. Feeds PS1/PM5/PM4/BP7 in B.11.
 #      (ncbiRefSeqSelect = single MANE transcript per gene; an intentional
 #      divergence from the evaSnp ncbiRefSeqCurated default — correct for a
 #      curated 8-gene panel. hg38 only; codes carry to hg19 with the variant
 #      universe via the per-track liftOver.)
 #
 # B.4  cmpVCEPRevel.py — REVEL PP3/BP4 thresholded scores
 #      Output: 22,466 features. REVEL >= 0.70 -> PP3_supporting; <= 0.40 ->
 #      BP4_supporting; in-between dropped (nothing drawn). fetch_revel_bedgraph()
 #      splits multi-bp bedGraph runs into 1-bp per-alt records (each position has
 #      its own REF allele).
 #
 # B.5  cmpVCEPCardioBoost.py — CardioBoost missense predictor (informational)
 #      Loads cm_prediction.RData via /usr/bin/Rscript; 31,236 missense predictions
 #      across the 8 genes. GRCh37 native -> add chr prefix -> liftOver to hg38
 #      (0 unmapped). Rendered as a default-OFF subtrack under the Bioinformatic
 #      composite (sibling to REVEL). Colored by CardioBoost's own published class
 #      (probability >= 0.9 Pathogenic, <= 0.1 Benign, else Indeterminate). NOT a
 #      CSpec-specified predictor and fires no ACMG code — informational only.
 #      name field includes ref/alt so same-aa variants via different nt are unique.
 #
 # B.7  cmpVCEPEvRepo.py — VCEP Curated Variants from EvRepo (Final, with codes)
 #      Output: 25 features (all MYH7). The complex repeat-notation p.Glu931del
 #      (c.2785GAG[2]) is recovered via a coords_via_hgvstovcf() fallback.
 #      Mouseover prefix "Final — VCEP EvRepo submission"; codesMet filtered to
 #      status="Met"; codesAll = Met + NotMet where enumerated. Standard 5-tier
 #      ACMG ramp.
 #
 # B.7c cmpVCEPWalsh2019.py — Walsh 2019 Pre-EvRepo curations (Table S6)
 #      Output: 155 features (3 rows outside our 8 genes filtered out). All 155
 #      are rendered: the 32 previously unmatched-to-ClinVar entries (complex
 #      indels + 4 TNNT2 SNVs incl. R92W) are placed via walsh_coords_via_tool()
 #      = hgvsToVcf on each gene's Walsh-paper transcript, gated on FILTER==PASS.
 #      TNNT2 uses NM_001001430.2 (MANE gives HgvsRefAssertedMismatch); TNNT2 hg19
 #      coords via liftOver from hg38. 27 variants tagged "Walsh-upgraded" (the new
 #      PM1 EF rule, asterisked in Table S6). Classification strings normalized to
 #      "Uncertain Significance" (not "VUS"). Standard ACMG ramp.
 #
 # B.8  cmpVCEPClinVar506161.py — VCEP ClinVar submissions (Final, no codes)
 #      Output: 199 features. Joined to variant_summary for hg38 + hg19 coords.
 #      All carry review status "reviewed by expert panel". 25 of 199 overlap
 #      EvRepo (flagged inline in the mouseover). Standard 5-tier ACMG ramp (the
 #      earlier desaturated palette was dropped). Carries no per-code evidence.
 #
 # B.9  cmpVCEPWalshOR.py — Walsh 2017 gene-level case-control OR (PS4)
+#      [RETIRED from the hub 2026-07 per the VCEP (Q5: do not display Walsh 2017/2019
+#       PS4; ORs use current cohorts). Script + data retained in the tree for reference.]
 #      Source: Walsh 2017 Supplementary_Tables_resubmit.xlsx, Tables S5A (HCM) /
 #      S5B (DCM). Output: 48 features = 8 genes x {HCM, DCM} x {All protein-
 #      altering, Truncating, Non-truncating}, each spanning the gene CDS (MANE),
 #      filterable by gene / cohortDisease / variantClass / ps4Strength.
 #      PS4 strength by OR 95%-CI lower bound (Strong >=20, Moderate >=10,
 #      Supporting >=5, else below threshold): Strong 1, Moderate 7, Supporting 9,
 #      below threshold 31. Standout: MYBPC3 truncating-HCM OR 118.8 (86.1-163.9)
 #      -> Strong. hg38 from MANE CDS; hg19 via liftOver. Replaces the earlier
 #      Walsh-2019 EF table (a different statistic; per CSpec, PS4 cites Walsh 2017).
 #
 # B.10 cmpVCEPAtlasEF.py — Atlas per-variant case-counts + UCSC-computed OR (PS4)
+#      [RETIRED from the hub 2026-07 per the VCEP (Q5). Script + data retained.]
 #      Source: cmp_downloads/atlas/variants/var_*.html (173 parsed; 16 parse
 #      failures). Output: 178 features (a variant renders once per disease cohort
 #      where it has data). UCSC computes the OR (Fisher 2x2 with Haldane 0.5
 #      correction; Woolf log-OR 95% CI) from Atlas case counts against ExAC
 #      controls; per-gene Walsh-2019 NonTrunc ExAC denominators are used for
 #      non-truncating vartypes, else the 60,706 baseline. PS4 strength binning:
 #      Strong 16, Moderate 30, Supporting 44, below threshold 88. Every mouseover
 #      carries the caveat that the OR is UCSC-computed against ExAC (not gnomAD).
 #
-# B.11 cmpVCEPProvisionalClass.py — Computable ACMG Criteria Summary
-#      (Formerly "NON-FINAL Provisional Classification"; reframed 2026-07-08 per
-#      the CM VCEP chair L. Bronicki. The track no longer computes an overall ACMG
-#      classification, only the computable codes that fire per variant.)
+# B.11 cmpVCEPProvisionalClass.py — Variant Evidence Summary
+#      (Formerly "NON-FINAL Provisional Classification"; reframed 2026-07 per the
+#      CM VCEP. Computes no overall classification. For each variant it lists,
+#      evidence-first, the computable evidence and the ACMG criterion each item
+#      would support.)
 #      Variant universe: identical to B.3 (10,974). Consumes the B.6 annotation
-#      TSV. Computable codes applied: BA1/BS1/PM2_supporting (B.3 FAF95),
+#      TSV. Evidence gathered: BA1/BS1/PM2_supporting (B.3 FAF95),
 #      PP3/BP4 (REVEL, missense), PM1 (CSpec hotspot region), PS1/PM5 (EvRepo
 #      P/LP reference, leave-one-out; a variant cannot earn the code from its
 #      own entry; PS1 reference excludes established splice-impact variants e.g.
 #      MYBPC3 c.2308G>A), PM4 (NMD-escaping truncating, non-MYBPC3: last exon or
 #      within 50 nt of the final exon-exon junction), BP7 (synonymous +
-#      SpliceAI < 0.20 + phyloP470way <= 0). PM1<->PM5 mutual exclusion enforced
-#      per CSpec (keep PM5, the variant-specific code; drop PM1; PM1+PS1
-#      co-occurrence flagged). HCM/DCM diseaseTag populated. A SpliceAI score
-#      >= 0.20 is recorded as an informational splice flag.
-#      Per-variant output = the list of triggered codes only; NO overall
+#      SpliceAI < 0.1, per Walker 2023 PMID 37352859; conservation requirement
+#      removed per the VCEP). PM1<->PM5 mutual exclusion enforced per CSpec (keep
+#      PM5, the variant-specific code; drop PM1; PM1+PS1 co-occurrence flagged).
+#      HCM/DCM diseaseTag populated. A SpliceAI score >= 0.20 is recorded as an
+#      informational splice flag.
+#      Mouseover is evidence-first ("evidence -> supports CODE"); NO overall
 #      classification is calculated. The GN002 combining logic remains in the
 #      script as classify() but is retired/uncalled. Clinical/functional codes
 #      (PS2/PS3/PS4/PP1/PP4/BS3/BS4) are not computed. Single neutral display
 #      color 91,107,122; no classification encoded.
-#      Output: 10,974 features. Code firing: BA1 210, BS1 367, PM2_Supporting
-#      9,893, PP3 1,436, BP4 1,647, BP7 1,356, PM1 1,293, PM4 27, PM5 11, PS1 0.
+#      Output: 10,974 features. Evidence firing: BA1 210, BS1 367, PM2_Supporting
+#      9,893, PP3 1,436, BP4 1,647, BP7 2,419, PM1 1,293, PM4 27, PM5 11, PS1 0.
 
 
 ##############################################################################
 # Phase C: Hub assembly
 ##############################################################################
 #
 # Files written/maintained by hand (NOT generated by the build scripts):
 #   hub.txt, genomes.txt
 #   cardiomyopathy.html        (shared description page; hub descriptionUrl)
 #   hg38/trackDb.txt
 #   hg19/trackDb.txt           (mirrors hg38; differs in bigDataUrl + an hg19
 #                              provenance header noting liftOver-derived coords)
 #
-# trackDb structure: 7 top-level groups -> 11 tracks. Three are composites:
-#   Bioinformatic (REVEL on + CardioBoost off), VCEP Curated Variants (EvRepo on
-#   + ClinVar off + Walsh 2019 off), PS4 Case-Control (Walsh OR on + Atlas EF off).
+# trackDb structure: 6 top-level groups -> 9 tracks. Two are composites:
+#   Bioinformatic (REVEL on + CardioBoost off) and VCEP Curated Variants (EvRepo on
+#   + ClinVar off + Walsh 2019 off). (The PS4 Case-Control composite was retired
+#   from the hub per the VCEP, 2026-07.)
 # bigBed filterValues on the ACMG-code / ps4Strength fields; default visibility
 # tuned so the first view is clean (dense for the big per-variant tracks).
 #
 # Validation:
 #   hubCheck https://hgwdev.gi.ucsc.edu/~lrnassar/track_hubs/cardiomyopathyVCEP/hub.txt
 #
 # Web access (sandbox; working dir served directly via symlink):
 #   ln -sf /hive/users/lrnassar/claude/RM37446 \
 #          /cluster/home/lrnassar/public_html/track_hubs/cardiomyopathyVCEP
 #   Hub URL: https://hgwdev.gi.ucsc.edu/~lrnassar/track_hubs/cardiomyopathyVCEP/hub.txt
 
 
 ##############################################################################
 # Phase D: Verification
 ##############################################################################
 #
 # (Build verification only. QA history, audit findings, and review/release
 #  readiness are tracked in Redmine #37446, not here.)
 #
 #   - hubCheck: silent pass (exit 0) on hg38 + hg19.
 #   - Cross-assembly parity (hg38 == hg19 feature counts):
 #       PM1 28 | PVS1 16 | AF 10,974 | REVEL 22,466 | EvRepo 25 | Walsh2019 155 |
-#       ClinVar 199 | WalshOR 48 | AtlasEF 178 | Computable codes 10,974 |
-#       CardioBoost 31,236
+#       ClinVar 199 | Variant evidence 10,974 | CardioBoost 31,236
 #   - Worked example, MYH7 p.Arg870His (NM_000257.4:c.2609G>A, chr14:23424839 hg38):
 #       EvRepo:       Pathogenic, codes PM1;PM2;PP1_Strong;PS4
 #       EvRepo R870C: Likely Pathogenic at the adjacent codon-870 position
 #                     (the PS1/PM5 partner)
 #       PM1:          within MYH7 167-931
 #       REVEL:        0.853 -> PP3_supporting
 #       gnomAD AF:    absent from v4.1 exomes -> PM2_supporting
-#       Walsh OR:     MYH7 non-truncating HCM, OR 12.0 (10.9-13.3) -> Moderate
-#       Computable codes: PM1+PM2+PP3 triggered; no overall classification is
-#                     computed (the VCEP Pathogenic call rests on PS4+PP1, manual)
+#       Variant evidence: PM1 hotspot, PM2 rarity, PP3/REVEL shown (each tagged with
+#                     the criterion it supports); no overall classification (the VCEP
+#                     Pathogenic call rests on PS4+PP1, manual)
 
 
 ##############################################################################
 # Phase E: Otto cron (TODO — gated on VCEP sign-off; do AFTER deployment)
 ##############################################################################
 #
 # DO NOT enable until the VCEP signs off (Phase F gate). The weekly refresh
 # touches EvRepo + ClinVar 506161 only; premature activation could surface
 # unreviewed VCEP curations on the public hub.
 #
 # Mirror the TP53 pattern at /hive/data/outside/otto/cardiomyopathyVCEP/
 # (doUpdate.sh + checkCMPVCEPClinVar.sh). Crontab (NOT activated until sign-off):
 #   # Cardiomyopathy VCEP weekly EvRepo + ClinVar update
 #   2x 03 * * 2 umask 002; /hive/data/outside/otto/cardiomyopathyVCEP/doUpdate.sh
 # Tuesday ~03:2x UTC — offset a few minutes from TP53 (03:15) and InSiGHT (03:10).
 
 
 ##############################################################################
 # Phase F: Deployment to hgdownload (TODO — gated on VCEP expert review)
 ##############################################################################
 #
 # Steps:
 #   1. (Done) MYH7 draft sent to Haley Garrett / Lucas Bronicki with a shared
 #      hgwdev session + interpretation questions.
 #   2. (Done 2026-07-08) Provisional track reframed to the Computable ACMG Criteria
 #      Summary per the CM VCEP chair: per-variant codes only, no overall classification.
 #   3. Symlink hub into hgdownload and coordinate autoPush:
 #        ln -sf /hive/users/lrnassar/claude/RM37446 \
 #               /usr/local/apache/htdocs-hgdownload/hubs/cardiomyopathyVCEP
 #      Public URL: https://hgdownload.soe.ucsc.edu/hubs/cardiomyopathyVCEP/hub.txt
 #   4. Commit to git (per CLAUDE.md, `refs #37446`):
 #        - This file -> ~/kent/src/hg/makeDb/doc/Cardiomyopathy.txt
 #        - All 12 build scripts -> ~/kent/src/hg/makeDb/scripts/cardiomyopathyVCEP/
 #      Then verify the cardiomyopathy.html GitHub "source code" links resolve and
 #      remove the PLACEHOLDER caveat; run encodeEmail.pl (already applied) and
 #      switch the Data Access section to the hgdownload (TP53-style) wording.
 
 
 ##############################################################################
 # Phase G: Recommended Track Set (TODO — after Phase E activation)
 ##############################################################################
 #
 # Create RTS sessions on dev (hgSession) for hg38 and hg19 with the hub loaded
 # at default subtrack visibility. Save under the VCEP folder alongside InSiGHT
 # and TP53.
 
 
 ##############################################################################
 # Phase H: Folder taxonomy (DEFERRED — until a 4th VCEP hub exists)
 ##############################################################################
 #
 # With InSiGHT + TP53 + Cardiomyopathy = 3 VCEP hubs, propose an RTS folder
 # structure when a 4th hub lands. Matches the TP53 makedoc Phase H deferral.
 
 
 ##############################################################################
 # References
 ##############################################################################
 #
 # Walsh R, Thomson KL, et al. (2017). Reassessment of Mendelian gene
 #   pathogenicity using 7,855 cardiomyopathy cases and 60,706 reference
 #   samples. Genet Med. PMID 27532257; DOI 10.1038/gim.2016.90.
 #
 # Walsh R, Mazzarotto F, et al. (2019). Quantitative approaches to variant
 #   classification increase the yield and precision of genetic testing in
 #   Mendelian diseases: the case of hypertrophic cardiomyopathy.
 #   Genome Medicine. PMID 30696458; DOI 10.1186/s13073-019-0616-z.
 #
 # Kelly MA, Caleshu C, et al. (2018). Adaptation and validation of the
 #   ACMG/AMP variant classification framework for MYH7-associated inherited
 #   cardiomyopathies. Genet Med. PMID 29300372.
 #
 # Jordan E, Peterson L, et al. (2021). Evidence-based assessment of genes
 #   in dilated cardiomyopathy. Circulation. PMID 33947203.
 #
 # Zhang X, Walsh R, et al. (2021). Disease-specific variant pathogenicity
 #   prediction significantly improves variant interpretation in inherited
 #   cardiac conditions (CardioBoost). Genome Medicine. PMID 33420041.
 #
 # Richards S, Aziz N, et al. (2015). Standards and guidelines for the
 #   interpretation of sequence variants. Genet Med. PMID 25741868.
 #
 # Remaining/deferred work is tracked in Redmine #37446 and
 # memory/rm37446_v2_punchlist.md (not duplicated here).