VARIANT LANDSCAPE

Where ASH1L variants occur—and what types they are.

Molecular-class heat maps show the distribution of loss-of-function, missense, and other positioned records across the 2,969-amino-acid canonical protein. Unrelated recurrence, cross-source matches, familial segregation, dual-variant configurations, public same-residue recurrence, and external provenance remain separate analytic layers.

MOLECULAR-CLASS HEAT MAP

Variant density by protein segment and variant type.

Each number is a positioned provenance record or cohort variant. Darker cells indicate more observations within that molecular class; color is not a severity scale.

External sourcebook222 positioned provenance rows · 200-aa segments
LoF / truncatingMissenseOther
Molecular class1–200201–400401–600601–800801–1,0001,001–1,2001,201–1,4001,401–1,6001,601–1,8001,801–2,0002,001–2,2002,201–2,4002,401–2,6002,601–2,8002,801–2,969
LoF / truncating1386341655294928
Missense1035527181513471117117
Other000000100000010
All positioned116131151135201861615261415
Connected cohort — privacy-protected49 LoF/missense people + 5 protein events from 3 complex/mixed people · broad segments
LoF / truncatingMissenseComplex / mixed
Molecular class1–499500–9991,000–1,4991,500–1,9992,000–2,4992,500–2,969
LoF / truncating1610680
Missense473112
Complex / mixed001103
All positioned51314895

External sourcebook: 222 positioned rows—85 LoF/truncating, 135 missense, and 2 other positioned records. Twelve additional external rows are off-scale or protein-position unavailable.

Connected cohort: the primary map contains 49 LoF/missense people—31 positioned LoF/truncating and 18 missense. A separate overlay contributes five protein-position events from three complex/mixed people: one in-frame deletion plus four consequences from the two dual configurations. That makes 54 plotted events, not 54 people. CNV/deletion, protein-undefined splice/noncoding, and pending records remain off the protein scale.

TWO DUAL CONFIGURATIONS · FOUR BROAD-SEGMENT EVENTS

Both dual configurations remain visible without exposing private coordinates.

Each row is one person. Each tile is one reported protein event placed in a 500-amino-acid segment. The two rows therefore represent two people and four plotted events—not four independent cases.

Dual missense

Two missense events reported in one person, shown only at broad-segment resolution.

01Missenseprotein segment 1,500–1,999
02Missenseprotein segment 2,500–2,969

Missense + nonsense

A missense and stop-gain event reported in one person, shown only at broad-segment resolution.

01Missenseprotein segment 2,500–2,969
02Stop-gainprotein segment 2,500–2,969

49primary protein events from 49 LoF/missense people

5complex/mixed protein events from 3 people

54plotted protein events

Why five complex/mixed events? The four broad-segment events above are joined by one connected-cohort in-frame deletion in protein segment 1,000–1,499. Exact connected-cohort coordinates are withheld. The missense-plus-nonsense row is retained as reported; the display does not infer phase, transcript mechanism, or two independent coding events.

RECURRENT & MULTI-VARIANT ARCHITECTURE

Independent recurrence, familial segregation, and dual-variant architecture are different signals.

The connected evidence includes two unrelated loss-of-function pairs, an additional internal-to-physician loss-of-function match, a five-person familial missense cluster, and two within-person dual-variant configurations. Exact private coordinates remain withheld.

2

unrelated LoF pairs

  • Same-stop frameshift pair: one female and one male, from unrelated families.
  • Truncating LoF pair: two people of the same sex, from unrelated families.
1

cross-source LoF match

One internally identified LoF case matches a physician-shared LoF case. The match remains visibly source-traced without exposing either private coordinate.

5

related missense cases

Five people belong to one familial missense cluster. This is one pedigree and one segregation structure—not five independent recurrence events.

2

dual-variant individuals

  • One person has a missense + missense configuration.
  • One person has a truncating + missense configuration.

READ THE STRUCTURE FIRST

Do not add unlike molecular events together.

The two unrelated LoF pairs can support independent recurrence comparison. The cross-source LoF match remains a separately traced evidence structure. The five missense cases support familial segregation and within-family variability. The dual findings require allele-specific inheritance, phase, and functional interpretation. These categories are not one prevalence denominator.

Open the longitudinal framework →

PUBLIC SAME-RESIDUE SIGNAL · AMINO ACID 1222

Two public frameshift consequences recur at one protein position.

Four retained public provenance rows resolve to two exact frameshift consequences at residue 1222. Connected-cohort and clinician-shared records remain in broad protein segments and are not asserted here as residue-level matches.

PUBLIC EXACT LAYER

2

frameshift consequences

  • p.Lys1222GlyfsTer10
  • p.Lys1222ArgfsTer11

Four retained public provenance rows point to these two consequences, so duplicated source rows do not become four independent people.

Trace the public source rows below →

EXTERNAL DATA SOURCES

Six source groups, one provenance registry.

The external landscape contains 234 retained provenance rows: 222 positioned and 12 off-scale or without a usable protein position. Repeated rows and cross-source overlap are retained as provenance and are not treated as independent people.

161154 positioned

SFARI Gene — ASH1L

Largest source bucket in the external molecular sourcebook.

3130 positioned

DECIPHER — ASH1L

Thirty-one provenance rows; DECIPHER also contributes 28 of the 50 structured external phenotype summaries.

1716 positioned

Literature / online sources

Published cases and other traceable online molecular sources retained as provenance rows.

85 positioned

Physician / clinician-shared

Restricted provenance rows retained in the totals but not exposed at exact-coordinate resolution.

66 positioned

ClinVar / GenomeConnect

Six public records with structured molecular provenance; selected rows include detailed phenotype summaries.

Privacy boundary

Public database and literature rows retain their public source resolution. Physician / clinician-shared rows and connected-cohort exact coordinates are not displayed. The cohort map remains aggregated into broad protein segments.