Blepharophimosis‑Impaired Intellectual Development Syndrome (BIS) — Comprehensive Research Report

Prepared: 2026‑08‑01 · Target entity: MONDO:0859139 · Gene: SMARCA2 (hgnc:11098) · Category: Mendelian, autosomal dominant, ultra‑rare

Verification note for curators. Quotes marked [V] were confirmed character‑for‑character against a cached PubMed abstract (references_cache/PMID_32694869.md, references_cache/PMID_38884529.md), against the open‑access Sarli et al. 2024 PDF read directly, or against a verbatim E‑utilities abstract fetch. Quotes marked [V?] were returned by a fetch layer that may have normalized whitespace — re‑run just fetch-reference and just validate-references before committing them as evidence snippet: values. Statements with no quote are attributed but not quotable.

NEC (Named Entity Confusion) preflight — this disease is HIGH RISK. "Blepharophimosis–intellectual disability syndrome" is a family of mechanistically unrelated MONDO entities: MONDO:0017393 (generic), MONDO:0011365 SBBYS type (KAT6B), MONDO:0010477 MKB type (MASP1), MONDO:0009583 Ohdo type, MONDO:0011432 Verloes type, MONDO:0979360 ADNP‑related, and MONDO:0859139 — the SMARCA2 entity that is the subject of this report. Any deep‑research output for this disease must be gene‑checked (SMARCA2) and OMIM‑checked (619293) before use. The MONDO↔OMIM↔ORPHA identity triple was confirmed in this session (see §1).

1. Disease Information

1.1 Overview

Blepharophimosis‑impaired intellectual development syndrome (BIS) is a congenital, recognizable neurodevelopmental/malformation syndrome defined in 2020 by Cappuccio et al. It is caused by de novo, non‑truncating (missense) heterozygous variants in SMARCA2 that cluster outside the ATPase/helicase domains, and it is clinically and molecularly distinct from the other SMARCA2 disorder, Nicolaides‑Baraitser syndrome (NCBRS, OMIM 601358), whose variants sit inside the ATPase domain.

[V] "Of 20 individuals, 14 showed a recognizable phenotype with recurrent features including epicanthal folds, blepharophimosis, and downturned nasal tip along with variable degree of intellectual disability (or blepharophimosis intellectual disability syndrome [BIS])." — Cappuccio et al., Genet Med 2020 (PMID:32694869; DOI 10.1038/s41436-020-0898-y)

[V] "We identified a novel recognizable syndrome named BIS associated with clustered de novo SMARCA2 variants outside the helicase domains, phenotypically and molecularly distinct from NCBRS." — PMID:32694869

[V] "Blepharophimosis with intellectual disability (BIS) is a recently recognized disorder distinct from Nicolaides-Baraister syndrome that presents with distinct facial features of blepharophimosis, developmental delay, and intellectual disability." — Sarli et al., Am J Med Genet C 2024 (PMID:38884529; DOI 10.1002/ajmg.c.32089)

The cardinal triad is blepharophimosis + epicanthus + developmental delay/intellectual disability, each present in 14/14 (100%) of the founding cohort, with a downturned/short nose, sparse eyebrows and eyelashes, thin upper lip vermilion, and broad nasal bridge completing a recognizable gestalt.

1.2 Identifiers

1.3 Synonyms

• BIS
• Blepharophimosis with intellectual disability
• Blepharophimosis intellectual disability syndrome
• SMARCA2‑related blepharophimosis‑intellectual disability syndrome
• Blepharophimosis‑impaired intellectual development syndrome

Do not use the bare string "blepharophimosis‑intellectual disability syndrome" as an unqualified synonym — it collides with ≥5 other MONDO entities (see NEC warning).

1.4 Provenance of information

All information is disease‑level aggregated (case series + ontology/database aggregation). There is no EHR‑derived or registry‑derived cohort for BIS. The evidence base is essentially two publications plus database curation:

1. Cappuccio et al. 2020 (PMID:32694869) — 20 individuals/18 families ascertained by NGS; 14 with the BIS gestalt. This is the source of the OMIM clinical synopsis and of essentially all HPO annotations.
2. Sarli et al. 2024 (PMID:38884529) — 15 individuals with BIS‑causing SMARCA2 variants (10 previously reported), 12 with class II ADNP HVDAS; episignature study.

[V] "The 15 individuals with de novo missense SMARCA2 variants all had the BIS phenotype and 10 of them were previously described (Cappuccio et al., 2020)." — Sarli et al. 2024, Results

Cumulative published BIS individuals as of mid‑2026: ≈19–20 (14 + 5 new in Sarli), plus 4 individuals labelled "BIS_atypical". This is a genuinely ultra‑rare, thinly reported entity.

2. Etiology

2.1 Primary cause

BIS is a monogenic, de novo dominant disorder. The cause is a heterozygous non‑truncating (missense) germline variant in SMARCA2, arising de novo in all reported cases.

[V] "In contrast to most NCBRS variants, all SMARCA2 variants associated with BIS are localized outside the helicase domains." — PMID:32694869

There is no known environmental, infectious, or multifactorial contribution. No teratogen, exposure, or maternal factor has been implicated.

2.2 Genetic risk factors

• Causal: de novo missense SMARCA2 variants in two hotspot regions (see §4).
• Susceptibility loci: none identified. No GWAS applies (ultra‑rare Mendelian).
• Modifier genes: none identified. Phenotypic variability (e.g., ambulation, speech, seizures) is unexplained; the founding cohort is too small to detect modifiers. Explicit knowledge gap.
• Recurrent alleles — the strongest genotype signal is recurrence of the same amino acid changes in unrelated probands, which is itself evidence of mutational hotspots at functionally constrained residues (Arg525, Arg937, Leu529, Gly513 — §4).

2.3 Environmental risk factors

None known. Advanced paternal age is a generic risk factor for de novo missense variants across NDDs and is biologically plausible here, but has not been specifically studied in BIS — do not curate it as a BIS risk factor without a BIS‑specific citation.

2.4 Protective factors

None known, genetic or environmental. Not applicable to a de novo dominant developmental disorder.

2.5 Gene–environment interactions

None reported. Not applicable.

3. Phenotypes

3.1 Frequency table — HPO annotations derived from OMIM:619293

Retrieved from the HPO/Monarch annotation API (ontology.jax.org/api/network/annotation/OMIM:619293). Denominators reflect Cappuccio's 14 BIS individuals (some features scored in subsets). These n/N fractions are the single best source for dismech frequency: bands, and each fraction is itself the quantitative justification the frequency‑evidence SOP requires.

Cardinal / obligate features (100%)

Craniofacial (the recognizable gestalt)

Curation note: sparse scalp hair is present in only 4/14 (29%) of BIS but ~97% of NCBRS — this is one of the two most discriminating features between the allelic disorders (the other being blepharophimosis, absent in NCBRS).

Nervous system / neurodevelopment

Eye / vision

Sarli Table 2 (BIS column, sourced from Cappuccio 2020) additionally records vision issue 64%, myopia/hypermetropia 42%, strabismus 14%. OMIM's clinical synopsis lists astigmatism, hyperopia, esotropia, myopia, and cortical visual impairment. HPO terms to consider: HP:0000545 Myopia, HP:0000540 Hypermetropia, HP:0000486 Strabismus, HP:0000642 Astigmatism, HP:0100704 Cerebral visual impairment (verify each with OAK before use — not verified in this session).

Limbs, skeleton, connective tissue

Contrast with NCBRS: prominent interphalangeal joints (84% in NCBRS) is not a BIS feature; BIS limb findings are contractures, tapering fingers, and foot deformity.

Growth (from Sarli Table 2, BIS column, "From Cappuccio et al. 2020")

HPO terms: HP:0000252 Microcephaly, HP:0001518 Small for gestational age, HP:0004322 Short stature (verify with OAK).

Other systems

Sarli Table 2 records GI/feeding issues 46%, abnormal ears 64%, hearing loss 0% (a useful negative — HPO "EXCLUDED" candidate for HP:0000365 Hearing impairment), brain abnormalities ~40%.

3.2 Phenotype characteristics

• Type mix: predominantly congenital physical/dysmorphic manifestations (craniofacial, ocular adnexal, limb) plus neurodevelopmental/behavioral features. No BIS‑specific laboratory abnormality exists — there is no biochemical marker. The only laboratory‑style biomarker is the peripheral‑blood DNA methylation episignature (§6, §10).
• Onset: congenital for the facial gestalt and structural anomalies; developmental delay evident in infancy. HPO onset term: HP:0003577 Congenital onset (verify).
• Severity: variable — ID ranges from mild to severe; 35% never develop speech ("absent speech", Sarli Table 2), 30% do not achieve independent walking (inferred from 70% "delayed walking" plus OMIM's "sometimes with inability to walk").
• Progression: non‑progressive / static developmental disorder. Unlike NCBRS, there is no report of progressive facial coarsening (coarse face only 17% in BIS vs ~80% in NCBRS) and no neurodegeneration. Seizures, when present, are not described as intractable in BIS. Musculoskeletal features (contractures, scoliosis) may progress mechanically with growth.
• Episodic elements: seizures (21%) and recurrent respiratory infections (~21–50%) are the episodic components.

3.3 Quality‑of‑life impact

No BIS‑specific QoL instrument data exist (no EQ‑5D, SF‑36, PROMIS, or disease‑specific PRO study). Explicit gap. Inferred per‑phenotype impact:

4. Genetic / Molecular Information

4.1 Causal gene

SMARCA2 (hgnc:11098; "SWI/SNF related BAF chromatin remodeling complex subunit ATPase 2"; aliases BRM, SNF2L2, hSNF2a, BAF190) at 9p24.3, 34 exons, RefSeq NM_003070.3, protein P51531, 1,590 aa. OMIM gene 600014.

SMARCA2 is one of the two mutually exclusive catalytic ATPase subunits of the mammalian SWI/SNF (mSWI/SNF, BAF) complex; the paralog is SMARCA4 (BRG1).

[V] "Nontruncating variants in SMARCA2, encoding a catalytic subunit of SWI/SNF chromatin remodeling complex, cause Nicolaides-Baraitser syndrome (NCBRS), a condition with intellectual disability and multiple congenital anomalies." — PMID:32694869

[V] "BIS is caused by pathogenic variants in SMARCA2, that encodes the catalytic subunit of the superfamily II helicase group of the BRG1 and BRM-associated factors (BAF) forming the BAF complex, a chromatin remodeling complex involved in transcriptional regulation." — PMID:38884529

4.2 Protein architecture (UniProt P51531, verified this session)

4.3 The two BIS variant clusters

Per OMIM 619293's molecular genetics narrative: the BIS mutations "clustered in 2 regions located outside of the catalytic ATPase helicase domains. Some occurred in exons 8 or 9, corresponding to a region between the HSA and helicase ATP-binding domain, whereas others occurred in exon 19, mapping to the linker region located between the DExx helicase ATP-binding and helicase C-terminal domains." [V?]

Mechanistic interpretation:

[V] "BIS-causing variants were found to cluster around an alpha-helix domain, that defines an interaction surface with other subunits of the BAF complex." — Sarli et al. 2024, Introduction

4.4 Reported BIS variants (NM_003070.3 / P51531)

From Sarli et al. 2024 Table 1 (read directly from the open‑access PDF; ACMG codes as published). This is the most complete public variant list for BIS.

Two curation cautions:

1. The p.(Asp510Gly) row's cDNA notation (c.6286C>A) is internally inconsistent with the protein change and with the 4,773‑nt CDS implied by a 1,590‑aa protein. Treat the cDNA as a probable typographical error in the published table; cite the protein change only, or omit.
2. p.(Met856Val) and p.(Leu766Val) fall inside UniProt's annotated helicase ATP‑binding domain (736–901), which is in tension with the paper's blanket statement that BIS variants are "outside the helicase domains." The discrepancy is a domain‑boundary definitional difference (the authors' "helicase domains" correspond to the ultra‑conserved catalytic motifs, not the full InterPro/UniProt DExx span). Curate the "outside the helicase domain" claim as the authors' framing, and do not assert that residue 856 lies outside UniProt's DExx domain. This is a legitimate, citable nuance — and it is precisely why p.(Leu766Val) is classified "atypical."

Hotspot summary: two clusters — ~486–534 (HSA C‑terminus / post‑HSA α‑helix, exons 8–9) and ~766–937 (ATPase lobe 1 and the lobe1–lobe2 linker, exon 19). Arg525 and Arg937 are recurrent, multi‑allelic hotspot residues (Arg→Cys and Arg→His at both), a classic signature of a constrained protein–protein interaction surface.

4.5 ClinVar landscape

ClinVar (db=clinvar, SMARCA2[gene] AND "blepharophimosis-impaired intellectual development syndrome"[dis]) returns 46 records. Selected germline classifications:

Critical interpretation caveat: the ClinVar "condition" field is submitter‑assigned, so any SMARCA2 variant submitted under the BIS condition label appears here — including benign variants and a multi‑exon deletion whose mechanism (haploinsufficiency) is inconsistent with the published BIS missense/hotspot model. Do not treat ClinVar condition labels as evidence that these variants cause BIS. Only p.(Arg525His), p.(Arg525Cys), p.(Glu512Lys), p.(Gly513Val), p.(Leu529Val), p.(Met856Val), p.(Arg937Cys/His) are literature‑anchored BIS alleles. Variants such as p.(Glu464Lys), p.(Asp534Gly), p.(Gln957Arg), p.(Pro625Leu), and p.(Tyr1489Cys) represent a plausible but unpublished phenotypic/allelic expansion — worth flagging as a KNOWLEDGE_GAP discussion, not as curated genetics.

4.6 Population allele frequency

All reported BIS alleles are absent from population databases (gnomAD/1000G/ExAC), consistent with ACMG PM2 applied throughout Sarli Table 1. I was unable to retrieve gnomAD constraint metrics (pLI/LOEUF/missense Z) in this session — the gnomAD GraphQL endpoint is not reachable via plain GET. Do not curate numeric constraint values from memory. Fetch them separately if needed.

4.7 Somatic vs germline

Germline, de novo. Not a somatic/cancer mechanism in BIS. Note for disambiguation: SMARCA2 is heavily studied in oncology as a synthetic‑lethal target in SMARCA4‑deficient cancers (PROTAC degraders — PMIDs 36357397, 36216795, 38557192, 39378885). That is mechanistically inverse to BIS and must not be conflated with BIS treatment (§12).

4.8 Functional consequence — loss of function, dominant negative, or gain of function?

The mechanism is not simple haploinsufficiency. Convergent lines of evidence:

1. ClinGen dosage sensitivity curation (SMARCA2, CCID:007899, last evaluated 2013‑03‑27): haploinsufficiency score 0 ("No Evidence for Haploinsufficiency"); triplosensitivity score 0. The curation notes that in the NCBRS/CSS papers "in neither paper is evidence provided for a loss of function effect of the mutation" [V?], that authors invoked a "gain of function effect of the mutant protein" [V?], that patients with 9p deletions overlapping SMARCA2 did not present with the NCBRS phenotype, and that SMARCA2 knockout mice are "viable and fertile."
2. All disease alleles are non‑truncating. Truncating SMARCA2 variants are not the established mechanism for either allelic disorder.
3. For NCBRS, the founding paper explicitly separates assembly from catalysis: [V] "The mutations cluster within sequences that encode ultra-conserved motifs in the catalytic ATPase region of the protein. These alterations likely do not impair SWI/SNF complex assembly but may be associated with disrupted ATPase activity." — Van Houdt et al., Nat Genet 2012 (PMID:22366787)
4. For BIS, the variant location implies the complementary defect: not catalysis but complex interface integrity — the residues sit on an α‑helix forming an interaction surface with other BAF subunits (Sarli 2024, quoted above). Cappuccio's yeast assay result supports a distinct functional class: [V] "Yeast phenotype assays differentiated NCBRS from non-NCBRS SMARCA2 variants."

Best current model: BIS alleles are hypomorphic/dominant‑interfering variants that perturb BAF module assembly or subunit engagement, whereas NCBRS alleles are ATPase‑catalysis‑impairing variants. Both are incorporated into complexes (hence dominant, non‑haploinsufficient behavior). Direct biochemical proof for the BIS class — complex co‑IP, nucleosome remodeling assays, ATAC‑seq in patient‑relevant cells — has not been published. This is the single largest mechanistic gap and is a strong candidate for a KNOWLEDGE_GAP discussion entry.

4.9 Modifier genes

None identified.

4.10 Epigenetic information

This is BIS's most distinctive molecular dimension, and unusually strong for so rare a disorder.

• BIS has its own peripheral‑blood DNA methylation episignature, distinguishable from NCBRS: [V] "Transcriptomic and DNA methylation signatures differentiated NCBRS from BIS and those with nonspecific phenotype." (PMID:32694869)
• BIS is a clinically available EpiSign™ episignature disorder — the London Health Sciences Centre/EpiSign conditions list includes the row "Blepharophimosis-impaired intellectual development syndrome (BIS) — SMARCA2 (619293)" under "Episignature Disorders", separately from "Nicolaides-Baraitser syndrome (NCBRS) — SMARCA2 (601358)" and "Helsmoortel-van der Aa syndrome (HVDAS) — ADNP (615873)." (verified by reading the EpiSign Conditions List PDF, June 2024 revision)
• BIS shares a phenotype‑specific episignature with class II HVDAS: [V] "A distinct episignature was shared by 15 individuals with BIS-causing SMARCA2 pathogenic variants and 12 individuals with class II HVDAS caused by truncating pathogenic ADNP variants." — PMID:38884529
• [V] "This represents first evidence of a sensitive phenotype-specific episignature biomarker shared across distinct genetic conditions that also exhibit unique gene-specific episignatures." — PMID:38884529
• Technical parameters (Sarli 2024, read from PDF): Illumina EPIC array; 31 individuals total (16 F/15 M); discovery n=23, validation n=4, non‑NCBRS‑non‑BIS n=4; 772,557 probes after filtering; 200 DMPs in initial selection, refined to a final set of 239 differentially methylated CpG probes; SVM classifier with MVP scores ~1 for cases; robustness by 20 rounds of leave‑25%‑out cross‑validation; the BIS‑HVDAS profile is relatively hypomethylated; highest DMP overlap with the HVDAS_C (ADNP central) cohort at 26%; of 4 "BIS_atypical" samples only 1 (case 31, p.Asn486Lys) clustered with the BIS‑HVDAS positives.
• Paralog convergence at the methylation level: an independent cohort found "common episignatures affecting homologous residues in highly conserved paralogous proteins (SMARCA2 M856V and SMARCA4 M866V)" [V] — Levy/Ciaccio et al., HGG Adv 2024 (PMID:38751117). This is direct evidence that the BIS Met856 allele and its SMARCA4 paralogous counterpart converge on one epigenomic consequence — excellent support for a shared‑mechanism pathophysiology node.

4.11 Chromosomal abnormalities

Not a mechanism for BIS. Distinguish carefully from 9p24.3/9p deletion syndromes, which delete SMARCA2 among many genes and produce a different phenotype (see PMIDs 41137173, 40196253, 41995485, 40836298 for contemporary 9p deletion cohorts). This distinction is itself the key argument against haploinsufficiency (§4.8).

5. Environmental Information

• Environmental factors: none known. No toxicant, radiation, pollutant, or occupational exposure implicated.
• Lifestyle factors: none. Not applicable to a de novo germline dominant disorder.
• Infectious agents: none causal. Recurrent bronchitis (21%) and pneumonia (7%) are secondary complications — plausibly related to hypotonia, GERD, and aspiration — not etiologic agents.

Section 5 is genuinely "not applicable" for BIS. Do not populate speculative environmental content.

6. Mechanism / Pathophysiology

6.1 Causal chain (proposed pathograph for dismech)

[MOLECULAR] De novo missense SMARCA2 variant in one of two hotspot clusters
    (HSA-adjacent α-helix ~486–534; ATPase lobe1/linker ~766–937)
    │
    ▼
[MOLECULAR] Altered SMARCA2 interaction surface with partner BAF subunits
    (α-helix at the inter-subunit interface; HSA-proximal Arp-module contact)
    → perturbed BAF (mSWI/SNF) complex assembly / subunit engagement,
      WITHOUT haploinsufficiency and WITHOUT the catalytic-motif lesion of NCBRS
    │
    ▼
[MOLECULAR] Impaired ATP-dependent nucleosome remodeling by BRM-containing BAF
    (GO:0140658 ATP-dependent chromatin remodeler activity ↓/ABNORMAL;
     GO:0006338 chromatin remodeling ABNORMAL)
    │
    ▼
[MOLECULAR] Altered chromatin accessibility at BAF-dependent enhancers/promoters
    → BIS-specific transcriptomic signature AND BIS-specific
      genome-wide DNA methylation episignature in peripheral blood
    │
    ▼
[CELLULAR]  Dysregulated BAF-dependent transcriptional programs in neural
    progenitors and differentiating neurons (npBAF→nBAF subunit switching;
    GO:0071564 npBAF complex)
    │
    ▼
[TISSUE]    Abnormal CNS development (thin corpus callosum, pontine hypoplasia,
    Chiari malformation in a minority; ~40% any brain abnormality)
    AND abnormal craniofacial / periocular morphogenesis
    (blepharophimosis, epicanthus, narrow palpebral fissures, nasal shape)
    │
    ▼
[ORGANISM]  Global developmental delay, intellectual disability, hypotonia,
    delayed/absent speech and ambulation, behavioral phenotype,
    recognizable facial gestalt, contractures/foot deformity,
    feeding difficulties/GERD, seizures (21%)

Convergent parallel arm (shared final common pathway, well supported):

Truncating ADNP variant in the bipartite nuclear localization (BNL) domain (class II HVDAS)
   → mutant ADNP retains/loses nuclear import but cannot recruit BAF
   → same disruption of BAF chromatin-remodeling output
   → SHARED phenotype-specific episignature + overlapping narrow-palpebral-fissure craniofacial phenotype

[V] "We speculate that class II ADNP variants and BIS-causing SMARCA2 variants disrupt the chromatin remodeling activity of the BAF complexes, thus explaining their functional overlap." — Sarli et al. 2024, Discussion

[V] "ADNP was found to bind directly to SMARCA2, SMARCA4, and SMARCC2 through its C-terminal end (Helsmoortel et al., 2014)." — Sarli et al. 2024, citing PMID reference to Helsmoortel et al. 2014

Sarli's Discussion also notes the mechanistic hypothesis that mutant ADNP "still binds to the DNA, but is no longer capable of recruiting the BAF complex, leading to diminished functionality of the complex and ultimately to deregulation of several genes" [V?].

6.2 Molecular pathways

• mSWI/SNF (BAF) ATP‑dependent chromatin remodeling — the sole primary pathway. Not a signaling cascade disorder (no Wnt/MAPK/mTOR/PI3K primary lesion). Reactome/KEGG coverage of SWI/SNF is thin; the authoritative mechanistic reviews are Nucleus 2023 "Mechanism of action of the SWI/SNF family complexes" (PMID:36633435) and the cryo‑EM structural literature.
• Downstream: BAF‑dependent transcriptional regulation by RNA polymerase II at enhancers; neural progenitor proliferation/differentiation programs.

6.3 Cellular processes

• Chromatin remodeling and nucleosome repositioning at BAF target loci.
• Neural progenitor self‑renewal → neuronal differentiation transition. BAF subunit switching is the canonical mechanism: [V?] "these subunits are replaced by the homologous BAF45b, BAF45c, and BAF53b" as progenitors differentiate, and blocking the switch impairs neuronal differentiation — Lessard et al., Neuron 2007 (PMID:17640523). UniProt's function annotation for P51531 likewise records participation in "neural development transitions between stem/progenitor and postmitotic states."
• Cell‑cycle control — BRM loss impairs G0/G1 arrest in response to confluency or DNA damage (Reyes et al. 1998, PMID:9843504). Relevance to BIS neurodevelopment is plausible but unproven.
• Not implicated: apoptosis‑centric, autophagy, inflammasome, or fibrotic mechanisms. No conforms_to fit with the fibrotic/senescence/hallmark modules.

6.4 Protein dysfunction

Altered protein–protein interaction surface rather than misfolding, aggregation, or catalytic‑site destruction. The BIS α‑helix cluster is predicted to sit at a BAF inter‑subunit interface. This places BIS squarely within the framework of Valencia et al.'s pan‑BAF structural analysis:

[V?] "Whereas mutations within the SMARCA2 helicase cause NCBRS, SMARCA2 mutations outside of this domain are implicated in a distinct disorder, blepharophimosis-impaired intellectual disability syndrome." — Valencia et al., Nat Genet 2023;55:1400–1412 (PMID:37500730, PMC10412456)

Valencia et al. establish that BAF‑complex genes "harbor the greatest number of de novo missense and protein-truncating variants among nuclear protein complexes" and that non‑truncating NDD variants "cluster in four key structural regions associated with high disease severity, including mSWI/SNF-nucleosome interfaces, the ATPase-core ARID-armadillo repeat (ARM) module insertion site, the Arp module and DNA-binding domains." [V?] The BIS HSA‑adjacent cluster is topologically consistent with the Arp‑module interface arm of that framework (the HSA helix engages ACTL6A/ACTB). Note: Valencia et al. is a structural/computational analysis — it reports no wet‑lab assay of the specific BIS alleles.

Newer BAF structural biology worth tracking for mechanism: a SWI/SNF‑specific Ig‑like domain ("SWIFT") acting as a transcription‑factor binding platform, Science 2026 (PMID:41477818).

6.5 Metabolic changes

None known. BIS is not a metabolic disorder; there is no enzyme deficiency, no accumulating metabolite, no metabolomic signature. Do not model against metabolic_intoxication_decompensation or lysosomal_substrate_accumulation.

6.6 Immune system involvement

No primary immune mechanism. Recurrent bronchitis/pneumonia are best explained by hypotonia/aspiration rather than immunodeficiency; no immunologic workup abnormality is reported. Do not curate an immunodeficiency claim.

6.7 Tissue damage mechanisms

Not a tissue‑injury disorder — the pathology is developmental (morphogenetic and neurodevelopmental), not degenerative. No oxidative stress, ischemia, fibrosis, or necrosis mechanism.

6.8 Biochemical abnormalities

None measurable clinically. No enzyme assay, receptor assay, or ion‑channel defect. The measurable molecular abnormality is epigenomic (methylation array), not biochemical.

6.9 Molecular profiling

6.10 Suggested ontology terms for the pathograph

Verified in this session via OLS4:

Plausible but NOT verified this session — run just validate-terms before use: GO:0006357 (regulation of transcription by RNA polymerase II), GO:0030182 (neuron differentiation), GO:0007399 (nervous system development), GO:0071565 (nBAF complex), CL:0000681 (radial glial cell — verified label in the CL search above), UBERON:0000955 (brain), UBERON:0002336 (corpus callosum), UBERON:0000988 (pons), CHEBI:15422 (ATP).

7. Anatomical Structures Affected

7.1 Organ level

Primary: - Central nervous system — global developmental delay/ID (100%), hypotonia (75%), seizures (21%), structural brain anomalies in ~40% (thin corpus callosum, pontine hypoplasia, Chiari malformation each ~1/14). - Periocular / eyelid and adnexal structures — blepharophimosis (100%), narrow palpebral fissures (71%), epicanthus (100%), ptosis (14%), sparse eyelashes (79%). - Craniofacial skeleton and soft tissue — nasal bridge/alae/tip, philtrum, upper lip vermilion, chin, forehead, ears (64%).

Secondary / variable: - Musculoskeletal — contractures (43%), tapering fingers (57%), pes planus (31%), talipes equinovarus (14%), scoliosis (25%), hip dysplasia (14%), toenail hypoplasia (36%). - Gastrointestinal — GERD (31%), feeding difficulties (15%). - Respiratory — recurrent bronchitis (21%), pneumonia (7%) (secondary). - Genitourinary — cryptorchidism (43% of males), hypospadias (14% of males), labial hypoplasia. - Cardiovascular — patent ductus arteriosus (1/13) — isolated report; not an established BIS feature. - Dentition — enamel hypoplasia, malocclusion, microdontia (each ~1/12 in BIS; note that dental disease is far more prominent in SMARCA2‑related epilepsy cohorts overall, 52.8%, which are ATPase‑domain/NCBRS‑weighted).

Body systems: nervous, visual/ocular adnexal, musculoskeletal, digestive, genitourinary, integumentary (hair/nails), respiratory (secondary).

Notably spared: hearing (0% hearing loss in BIS vs 32% in HVDAS — a useful discriminator); no hepatic, renal‑parenchymal, endocrine, or hematologic involvement reported.

7.2 Tissue and cell level

Because the lesion is in a broadly expressed chromatin remodeler, the affected cell populations are inferred from BAF developmental biology rather than measured in BIS tissue:

• Neural progenitor cells / radial glia (CL:0011020, CL:0000681) — npBAF‑dependent proliferation.
• Postmitotic differentiating neurons — nBAF‑dependent maturation, dendritic outgrowth.
• Cranial neural crest–derived mesenchyme and periocular mesenchyme — the presumed substrate for the blepharophimosis/epicanthus/nasal phenotype. This is an inference: no BIS study has examined periocular or neural‑crest tissue. Flag as a HUMAN_MODEL_MISMATCH/KNOWLEDGE_GAP rather than asserting it.
• Eyelid epithelium and tarsal plate (UBERON:0003844/0003845, UBERON:0004773/0004774) — anatomically implicated by the phenotype.

7.3 Subcellular level

• Nucleus (primary compartment; UniProt: nucleus, also localizes to sites of DNA damage).
• Chromatin / nucleosome — the functional substrate.
• SWI/SNF superfamily‑type complex (GO:0070603) and npBAF complex (GO:0071564) as the molecular machines.

7.4 Localization and lateralization

Bilateral and symmetric. Blepharophimosis, epicanthus, narrow palpebral fissures, contractures, and foot deformities are described bilaterally. Plagiocephaly (21%) is the only asymmetric craniofacial finding recorded. No lateralized/unilateral pattern is reported.

8. Temporal Development

8.1 Onset

• Congenital. Facial gestalt, epicanthus, blepharophimosis, and structural anomalies are present at birth. Low birth weight (36%) and short birth length (29%) indicate prenatal onset of the growth phenotype.
• Onset pattern: insidious recognition of the neurodevelopmental component during infancy; delayed walking and delayed speech become apparent in the first 1–3 years.
• Prenatal detectability: limited. No BIS‑specific prenatal ultrasound signature exists. For orientation only, a fetus with a de novo SMARCA2 pathogenic variant presented with caudal regression, sacral agenesis, and congenital vertical talus (PMID:38877377, Prenat Diagn 2024) — but that report frames the findings as an expansion of the NCBRS spectrum, not BIS. Do not attribute caudal regression to BIS.

8.2 Progression

• Course: static / non‑progressive developmental disorder with a chronic, lifelong duration.
• Stages: no formal staging system exists or is appropriate.
• Rate: not applicable — developmental trajectory, not degeneration.
• Ages reported: Sarli's BIS subjects spanned 1 to 17.5 years; the oldest published individuals are adolescents/young adults. There is no adult natural‑history data at all. Whether the BIS facial gestalt evolves with age (as NCBRS coarsening does) is unknown — a concrete, high‑value gap, and a point of practical diagnostic importance since gestalt recognition drives ascertainment.
• Distinguishing feature vs NCBRS: BIS shows only 17% coarse facies vs ~80% in NCBRS, and no reported progressive coarsening; BIS seizures (21%) are neither as frequent nor as refractory as NCBRS epilepsy (65%, "often difficult to manage" per GeneReviews).

8.3 Patterns

• Remission: none; not applicable.
• Critical periods: (a) embryonic/fetal craniofacial and CNS morphogenesis — the window in which the malformation phenotype is determined and therefore not therapeutically accessible postnatally; (b) infancy–early childhood — the actionable window for early intervention, feeding/GERD management, and amblyopia prevention; (c) growth years — contracture, scoliosis, and foot‑deformity surveillance.

9. Inheritance and Population

9.1 Epidemiology

• Prevalence: no published estimate. The Orphanet epidemiology endpoint returned no prevalence record for ORPHA:637013 in this session; MONDO/OMIM carry none.
• Reported cases: ≈19–20 published individuals (14 in Cappuccio 2020 + 5 new in Sarli 2024), plus 4 "BIS_atypical". By any reasonable reading this is ultra‑rare, i.e. CASES_IN_LITERATURE with prevalence_class: NOT_YET_DOCUMENTED or ULTRA_RARE in dismech terms. Do not fabricate a numeric rate.
• For calibration, the allelic disorder is itself vanishingly rare: [V?] "The prevalence of SMARCA2-NCBRS is not known, but is estimated to be extremely low. Fewer than 100 affected individuals have been described." — GeneReviews, SMARCA2‑Related Nicolaides‑Baraitser Syndrome (NBK321516)
• Incidence: unknown; not estimable.
• Ascertainment bias: BIS is diagnosed by exome/genome sequencing in syndromic NDD cohorts, then gestalt‑confirmed. It is almost certainly under‑diagnosed, because (i) the gestalt was only described in 2020, (ii) non‑ATPase‑domain SMARCA2 missense variants would previously have been called VUS, and (iii) episignature testing that resolves them is not universally available.

9.2 Inheritance genetics

9.3 Population demographics

• Ethnic/geographic distribution: The founding cohort was ascertained across Italy, France, UK, Portugal, Czech Republic, Netherlands, Japan, Canada, USA; Sarli added Turin, Milan, Rome, Caen, Dijon, Coimbra, Birmingham, Manchester, Oxford, Tokyo, and Greenwood SC. There is no evidence of ethnic or geographic clustering — the distribution reflects the geography of the collaborating clinical‑genetics networks, not disease biology.
• Variant geography: none. Recurrent alleles (Arg525, Arg937) recur across unrelated families in different countries, consistent with independent de novo mutational hotspots rather than shared ancestry.
• Sex ratio: approximately balanced. Sarli's 15 BIS individuals were 8 female / 7 male; the founding HPO denominators imply 7 males and 5 females scored for sex‑specific features. No sex bias; no X‑linked or sex‑limited component.
• Age distribution of affected individuals: children and adolescents (1–17.5 years reported). Adults are absent from the literature — an ascertainment artifact of a 2020 disease definition, not evidence of reduced survival.

10. Diagnostics

10.1 Diagnostic strategy (recommended pathway)

1. Clinical recognition of the gestalt: blepharophimosis + epicanthus + narrow palpebral fissures + sparse eyebrows/eyelashes + broad nasal bridge + short nose with downturned tip + thin upper lip + DD/ID.
2. Exome or genome sequencing (or a comprehensive ID/NDD gene panel including SMARCA2) — the route used in the founding cohort: [V] "By next-generation sequencing, we identified candidate variants in SMARCA2 in 20 individuals from 18 families with a syndromic neurodevelopmental disorder not consistent with NCBRS."
3. Variant localization — determine whether the missense variant falls in a BIS hotspot (outside the catalytic ATPase motifs, in the HSA‑adjacent α‑helix ~486–534 or the lobe1/linker region ~766–937) vs an NCBRS ATPase‑motif position. Domain position is the primary genotype–phenotype discriminator.
4. Trio testing to establish de novo status (ACMG PS2 — applied to every BIS variant in Sarli Table 1).
5. Episignature testing (EpiSign™) for VUS resolution and confirmation. BIS is an explicitly listed EpiSign episignature disorder (SMARCA2, OMIM 619293), distinct from the NCBRS entry.

10.2 Genetic testing modalities

10.3 Omics‑based diagnostics

• Epigenomics — clinically actionable. Genome‑wide methylation (Illumina EPIC) episignature analysis, delivered as EpiSign™, is the one omics assay in routine diagnostic use for BIS. It resolves SMARCA2 VUS and separates BIS from NCBRS. Caveat to curate: BIS and class II HVDAS share the phenotype‑specific episignature (PMID:38884529), so a positive BIS‑HVDAS signature is not by itself gene‑specific — it must be interpreted alongside the sequencing result.
• Transcriptomics — used as a research‑grade discriminator in the founding study (blood leukocytes); not a routine clinical test.
• Proteomics / metabolomics / liquid biopsy — no role.

10.4 Variant‑interpretation tooling

A gene‑specific machine‑learning pathogenicity predictor has been developed using SMARCA2 and SMARCA4 NDD variants — reported accuracy 0.93 on holdout data for the gene‑specific model and 0.91 for the generalized BAF‑subunit predictor, outperforming generic tools; the authors emphasize "gene-specific calibration of predictors" (Reilly et al., HGG Adv 2026, PMID:41764075; preprint PMID:41000737). This is directly relevant to reclassifying the numerous SMARCA2 VUS submitted under the BIS condition label (§4.5).

10.5 Non‑genetic clinical tests

There is no diagnostic biochemical, imaging, or electrophysiologic test for BIS. The following are management/complication assessments, not diagnostics:

10.6 Clinical criteria and differential diagnosis

No formal consensus diagnostic criteria exist for BIS (as is also true for NCBRS per GeneReviews). Diagnosis = suggestive gestalt + de novo non‑truncating SMARCA2 variant at a hotspot position, ideally with episignature support.

Differential diagnosis — the discriminating features that matter:

10.7 Screening

• Newborn screening: not applicable and not appropriate (no treatable metabolic component; no NBS assay).
• Carrier screening: not applicable (de novo dominant).
• Cascade screening: not applicable for de novo variants; would apply only to the offspring of an affected individual.
• Reverse‑phenotyping screening: the practically useful "screen" is retrospective re‑analysis of unsolved NDD cohorts for non‑ATPase‑domain SMARCA2 missense variants, followed by episignature confirmation — the approach that generated the entity in the first place.

11. Outcome / Prognosis

11.1 Survival and mortality

• No mortality data exist for BIS. No deaths are reported among the ~20 published individuals.
• Life expectancy: unknown, presumed not markedly shortened in the absence of severe epilepsy or major organ malformation. There is no cardiac, renal, hepatic, or oncologic component to drive excess mortality. Recurrent respiratory infections (21–50%) and aspiration risk from feeding difficulty/GERD (46%) are the plausible mortality contributors in the most severely affected, non‑ambulant subset — by analogy with severe NDDs generally, not from BIS data.
• 5‑/10‑year survival, disease‑specific mortality: not applicable / not measured. Do not populate numeric survival fields.

11.2 Morbidity and function

• Dominant morbidity is neurodevelopmental disability: ID 100%; absent speech 35%; non‑ambulant ~30%; behavioral problems 65%. Most affected individuals will require lifelong support, with a substantial minority (those with absent speech and no independent ambulation) requiring full‑time care.
• Secondary morbidity: orthopedic (contractures 43%, scoliosis 25%), nutritional/GI (46%), ophthalmologic (64%), respiratory infections.
• Quality‑of‑life instruments: none administered. ICF‑based functional outcome data: none. Explicit gap.

11.3 Disease course and complications

Chronic, stable, lifelong. Complications to anticipate: aspiration pneumonia, failure to thrive, amblyopia/uncorrected refractive error, progressive contracture and scoliosis with growth, seizure emergence, dental disease, behavioral escalation in adolescence. Recovery potential: none — this is a static developmental disorder. Function improves with development and intervention; the underlying chromatin lesion is not reversible.

11.4 Prognostic prediction

• Best candidate prognostic factor: variant position. In the allelic disorder this is established — [V?] "All individuals with a pathogenic variant within the C-terminal helicase region of the ATPase domain have severe intellectual disability and epilepsy, a frequency higher than that in individuals with pathogenic variants in other parts of the gene." (GeneReviews NBK321516, NCBRS). For BIS specifically, no intra‑BIS genotype–severity correlation has been established — indeed three individuals sharing p.Arg937His differ phenotypically, arguing that variant identity alone is insufficient. Curate this as an open question, not as a correlation.
• Prognostic biomarkers: none. The episignature is diagnostic, not prognostic — MVP score has no demonstrated relationship to severity.
• Practical early prognostic indicators (severity of hypotonia, age at walking, speech acquisition by age 3, seizure onset) are generic NDD predictors, not BIS‑validated.

12. Treatment

There is no disease‑modifying or targeted therapy for BIS. Management is entirely supportive, anticipatory, and multidisciplinary. No published BIS‑specific management guideline exists; the closest applicable framework is the GeneReviews management section for SMARCA2‑related NCBRS, which states that no cure exists and care is supportive and multidisciplinary.

12.1 Supportive / rehabilitative management (the core of care)

NCIT IDs above are drawn from the dismech‑vetted list in CLAUDE.md; verify NCIT:C15315, C49236, C181743 with OAK before use.

12.2 Pharmacotherapy specifics

• Anti‑seizure medications: no BIS‑specific evidence. Extrapolatable data: in a SMARCA2‑related epilepsy cohort (24 own + 46 literature cases, mostly ATPase‑domain variants), seizure control was achieved in 45.8% using valproate or levetiracetam (PMID:40513420, Seizure 2025). GeneReviews likewise notes levetiracetam and valproic acid often show initial response in NCBRS. Curate this as extrapolated, NCBRS/ATPase‑weighted evidence, not BIS evidence — BIS epilepsy is both rarer (21%) and apparently milder.
• Pharmacogenomics: no BIS‑specific PGx. Standard CPIC guidance applies to the drugs used (e.g., HLA‑B*15:02/carbamazepine, CYP2C9/phenytoin) — general practice, not disease‑specific.

12.3 Advanced therapeutics — status: none, and one important negative

• Gene therapy / gene editing: none. Conceptually challenging: the phenotype is largely prenatal and morphogenetic (craniofacial structure is fixed before birth), and the mechanism is a dominant‑interfering missense allele in a broadly required chromatin regulator — allele‑specific silencing (ASO/siRNA against the mutant allele) is the only theoretically coherent modality, and no such program exists.
• ASOs / siRNA / mRNA: none. Note SMARCA2 has no ASO program; the mechanism does not fit any of the three FDA‑approved ASO paradigms cleanly (there is no splice lesion and no accumulating toxic transcript in the usual sense).
• Cell therapy, immunotherapy: not applicable.
• ⚠ Critical negative to record — SMARCA2 degraders are NOT a BIS therapy. Multiple clinical programs target SMARCA2 for degradation in SMARCA4‑deficient cancers (NCT06560645 PRT7732, terminated; NCT05639751 PRT3789, completed; NCT06682806, terminated; NCT06561685 LY4050784; NCT07284186 PLX‑61639; preclinical PMIDs 36357397, 36216795, 38557192, 39378885, 40280558, 41184243). These reduce SMARCA2 and would be expected to be harmful, not helpful, in a SMARCA2‑opathy. An automated pipeline mining "SMARCA2 + drug" will surface these; they must be explicitly excluded from BIS treatment curation.

12.4 Trials and registries

No interventional trial for BIS exists. Relevant observational studies that may enroll BIS individuals: - NCT01793168 — Rare Disease Patient Registry & Natural History Study (CoRDS, Sanford), recruiting, observational. - NCT04463316 — "GROWing Up With Rare GENEtic Syndromes", recruiting, observational.

12.5 Treatment outcomes, adverse events, algorithms

• Response rates: none reported (no trials).
• Adverse events: those of the individual supportive interventions; nothing BIS‑specific.
• Algorithms / combination therapy / genotype‑guided treatment: none exist. Explicit gap — a BIS management guideline is an unmet need, and the NCBRS GeneReviews surveillance schedule (growth, seizures, development at each visit; dental review at least every 6 months) is the best available proxy.

13. Prevention

13.1 Prevention levels

• Primary prevention: not possible. De novo germline missense variants are not preventable. No modifiable exposure exists.
• Secondary prevention (early detection and intervention): the actionable level. Early molecular diagnosis via ES/GS in infants with DD + blepharophimosis enables early intervention, feeding/GERD management, amblyopia prevention, and avoids the diagnostic odyssey. Episignature testing shortens time‑to‑diagnosis for SMARCA2 VUS.
• Tertiary prevention (complication avoidance): the practical mainstay — aspiration prevention (feeding/GERD management), amblyopia prevention (refractive correction, ptosis surgery where the visual axis is obstructed), contracture and scoliosis prevention (PT, orthoses, positioning), seizure control, dental prophylaxis, nutritional optimization.

13.2 Immunization

No disease‑specific vaccine. Routine immunization is important and should be emphasized, given recurrent bronchitis (21%) and pneumonia (7%) — including influenza and pneumococcal vaccination per standard schedules for children with neurodisability.

13.3 Screening and reproductive options

• Population screening: none; not justified for an ultra‑rare de novo disorder.
• Genetic counseling: indicated for every family. Key messages: (i) the variant is de novo, (ii) sibling recurrence risk is very low (residual risk for undetected parental gonadal mosaicism, conventionally ~1%), (iii) offspring risk for an affected individual is 50%.
• Prenatal testing / PGT: technically available for a known familial variant (relevant only in the theoretical case of a transmitting affected parent, or for reassurance in a subsequent pregnancy). PGT‑M is not indicated for a proven de novo variant other than at parental request.
• Risk stratification: not applicable.

13.4 Behavioral, public health, environmental interventions, prophylaxis

Not applicable to etiology. The only public‑health‑relevant lever is access to exome/genome sequencing and to episignature testing in NDD diagnostic pathways — a health‑systems issue, not an environmental one. No prophylactic medication is indicated.

14. Other Species / Natural Disease

• Taxonomy: Homo sapiens (NCBITaxon:9606) is the only species in which BIS occurs. BIS is a human clinical entity defined by a human gestalt.
• Breed (VBO): not applicable. No breed‑associated SMARCA2 disorder is known.
• Orthologous genes: mouse Smarca2 (MGI:99603, verified via MGI), with orthologs across vertebrates (rat, zebrafish smarca2), and the deeply conserved yeast ortholog SNF2 in Saccharomyces cerevisiae — the ortholog that made the founding functional assay possible. Retrieve NCBI Gene IDs for non‑human orthologs before curating them; I verified only the mouse MGI ID in this session.
• Naturally occurring animal disease: none reported. No OMIA entry for a SMARCA2 Mendelian disorder in a domestic species. BIS has no natural animal counterpart.
• Veterinary relevance: none.
• Comparative pathology: the informative comparison is negative and mechanistically important — mouse Smarca2/Brm null animals do not model BIS or NCBRS: [V?] "BRM-/- mice develop normally, suggesting that an observed up-regulation of the BRG1 protein can functionally replace BRM." (Reyes et al., EMBO J 1998, PMID:9843504). ClinGen's dosage curation cites exactly this — SMARCA2 knockout mice are "viable and fertile" — as evidence against a loss‑of‑function mechanism.
• Evolutionary conservation: very high. The ATPase/helicase motifs are described as "ultra-conserved" (Van Houdt 2012 [V]), and BAF/SWI‑SNF architecture is conserved from yeast Snf2 to human BRM/BRG1 — the basis for the yeast complementation strategy. Separately, the human paralogs SMARCA2/SMARCA4 converge at homologous residues (M856V/M866V, shared episignature; PMID:38751117), evidence of conserved residue‑level function within the paralog pair.
• Zoonotic potential / cross‑species susceptibility: not applicable (genetic disorder).

15. Model Organisms

15.1 Yeast (Saccharomyces cerevisiae) — the model that defined BIS

The only model system with published, BIS‑allele‑specific data. Cappuccio et al. expressed the human variants in the context of the yeast Snf2 ortholog and used growth/phenotype assays as a functional stratifier:

[V] "To stratify variant interpretation, we functionally analyzed SMARCA2 variants in yeasts and performed transcriptomic and genome methylation analyses on blood leukocytes."

[V] "Yeast phenotype assays differentiated NCBRS from non-NCBRS SMARCA2 variants."

Structural inference from the yeast Snf2 homolog placed the mutated BIS residues on an α‑helix at the SWI/SNF inter‑subunit interface [V] (Sarli 2024 Introduction, summarizing Cappuccio 2020).

• Strengths: deep conservation of the Snf2 ATPase; tractable, quantitative, scalable; directly discriminated the two disease classes — exactly the assay a variant‑classification pipeline needs.
• Limitations: yeast has no nervous system, no eyelids, no craniofacial development, and lacks the metazoan BAF subunit repertoire (no ADNP, no BAF45/53 paralog switching, no cBAF/PBAF/ncBAF diversification). It can report on catalysis and complex integrity, not on phenotype. Curate yeast results as evidence_source: IN_VITRO (or OTHER for a heterologous‑organism assay) — not as evidence for human phenotypes.

15.2 Mouse (Mus musculus) — available but non‑recapitulating

• Gene: Smarca2, MGI:99603. 10 mutant alleles across endonuclease‑mediated (5), targeted (4), and gene‑trapped (1) categories (MGI, verified).
• Reported mutant phenotype systems: behavior/neurological, growth/body size, homeostasis/metabolism, muscle physiology, reproductive system. MGI notes homozygotes for a targeted mutation "may exhibit infertility and a slightly increased body weight in some genetic backgrounds."
• Founding characterization (Reyes et al. 1998, PMID:9843504): null mice develop normally with BRG1 upregulation compensating; adults ~15% heavier; increased hepatic mitotic index; mutant embryonic fibroblasts [V?] "deficient...ability to arrest in the G0/G1 phase of the cell cycle in response to cell confluency or DNA damage."
• Phenotype recapitulation: poor to absent. No mouse shows blepharophimosis, ID, or the BIS craniofacial gestalt. No knock‑in mouse carrying a BIS hotspot allele (e.g., Arg525His, Arg937His, Met856Val) has been reported.
• This is a textbook HUMAN_MODEL_MISMATCH, not a KNOWLEDGE_GAP: model‑organism evidence exists (Smarca2‑null mice), but its translational validity is the open question — because the null does not model a dominant‑interfering missense allele, and BRG1 compensation in mouse may not mirror human BRM/BRG1 dosage relationships. The obvious resolving experiment is a conditional knock‑in of a BIS hotspot allele with craniofacial and neurodevelopmental phenotyping.

15.3 Human cellular models

• Patient peripheral blood leukocytes — the substrate for both the BIS transcriptomic and DNA methylation signatures (Cappuccio 2020; Sarli 2024). Accessible, clinically validated, but not the affected tissue — an inherent limitation: a blood episignature is a surrogate readout of chromatin dysregulation, not a measure of neural or periocular pathobiology.
• iPSC‑derived neurons / neural progenitors / organoids: none reported for BIS. This is the highest‑value missing model. A BAF‑relevant precedent exists in the MorPhiC framework (iPSC‑derived multicellular systems, null alleles, category: Cellular phenotypes, evidence_source: IN_VITRO), but SMARCA2 is not among the flagged MorPhiC anchor genes (ISL1, EOMES, GCM1, NKX2‑1) and no MorPhiC SMARCA2 dataset was identified.
• Cell lines / biochemical reconstitution: extensive SMARCA2 biochemistry exists in the cancer literature (degrader pharmacology, SMARCA4‑deficient synthetic lethality). None of it interrogates BIS alleles. Reconstituted BAF complexes with BIS variants — testing assembly, ACTL6A/Arp‑module engagement, nucleosome remodeling, and ATPase activity — is the missing biochemistry.

15.4 Zebrafish, Drosophila, C. elegans

• Zebrafish smarca2 ortholog exists; no BIS model reported. Zebrafish would be a reasonable system for craniofacial/neural‑crest patterning readouts.
• Drosophila brm is the founding family member and the source of the name "brahma"; classic developmental genetics, but no BIS allele modeling.
• C. elegans — not used.

15.5 Model resources

MGI (informatics.jax.org, MGI:99603), IMPC/KOMP/IMSR for Smarca2 allele availability, ZFIN for smarca2, FlyBase for brm, SGD for SNF2, Alliance of Genome Resources for cross‑species integration, Cellosaurus/ATCC for cell lines.

15.6 Research applications and the honest bottom line

Existing models can address: BAF catalytic mechanism (yeast, biochemistry), BAF subunit switching in neural development (mouse, Lessard 2007), and variant classification (yeast assay, ML predictors). No existing model addresses the two questions that matter most for BIS: why these specific interface variants cause blepharophimosis, and what the dominant‑interfering allele does to BAF function in human neural and periocular tissue.

Appendix A — Reference list with verification status

Appendix B — Prioritized knowledge gaps (candidate discussions entries)

1. KNOWLEDGE_GAP — mechanism of the BIS allele class. No biochemical assay of BIS variants on BAF assembly, Arp‑module engagement, nucleosome remodeling, or ATPase activity. attaches_to the interface/remodeling nodes. Proposed experiments: co‑IP/mass spec of BAF from BIS‑variant cells; in vitro reconstitution + remodeling assay; ATAC‑seq/CUT&RUN in isogenic knock‑ins.
2. KNOWLEDGE_GAP — why blepharophimosis? No mechanistic account links SMARCA2 interface variants to periocular/eyelid morphogenesis. Nothing is known about BRM function in periocular neural crest.
3. HUMAN_MODEL_MISMATCH — mouse Smarca2 null is viable, fertile, and normal‑appearing, so it cannot model a dominant‑interfering missense allele; BRG1 compensation may not translate. Resolving experiment: BIS hotspot knock‑in mouse.
4. KNOWLEDGE_GAP — no adult natural history. All reported individuals are ≤17.5 years. Unknown whether the gestalt evolves, whether new complications emerge, and what adult function/QoL look like.
5. KNOWLEDGE_GAP — intra‑BIS genotype–phenotype correlation. Three individuals share p.Arg937His yet differ phenotypically; no severity predictor exists, and no modifier has been sought.
6. KNOWLEDGE_GAP — allelic/phenotypic expansion outside ClinVar's literature. Several SMARCA2 variants classified Likely pathogenic for BIS in ClinVar (p.Glu464Lys, p.Asp534Gly, p.Gln957Arg, p.Pro625Leu, p.Tyr1489Cys) sit outside the two published clusters and are unpublished. Do the clusters need redefining, or are these mislabeled?
7. Data‑quality flags: (a) the p.(Asp510Gly)/c.6286C>A inconsistency in Sarli Table 1; (b) UniProt places p.Met856Val and p.Leu766Val inside the annotated helicase ATP‑binding domain (736–901), in tension with "outside the helicase domains"; (c) two UMLS CUIs (C5443984, C5816784) map to MONDO:0859139.
8. Diagnostic caveat worth curating explicitly: the BIS episignature is phenotype‑specific, not gene‑specific — it is shared with class II ADNP HVDAS, so a positive result requires sequencing to assign the gene.

Sources: - Cappuccio et al. 2020, Genetics in Medicine (PMID:32694869) · publisher - Sarli et al. 2024, Am J Med Genet C (PMID:38884529) · open-access PDF - Van Houdt et al. 2012, Nature Genetics (PMID:22366787) - Valencia et al. 2023, Nature Genetics (PMID:37500730 / PMC10412456) - DNA methylation analysis in NDDs, HGG Advances 2024 (PMID:38751117) - Genotype and phenotype correlation in epilepsy patients with SMARCA2 variants, Seizure 2025 (PMID:40513420) - Lessard et al. 2007, Neuron (PMID:17640523) - Reyes et al. 1998, EMBO J (PMID:9843504) - SMARCA2-Related Nicolaides-Baraitser Syndrome, GeneReviews (NBK321516) - OMIM #619293 · MedGen C5443984 · Monarch MONDO:0859139 - Orphadata cross-referencing API, ORPHA:637013 - HPO annotation API, OMIM:619293 - UniProt P51531 · HGNC:11098 · MGI:99603 - ClinGen Dosage Sensitivity, SMARCA2 - EpiSign Conditions List (June 2024) - ClinicalTrials.gov API v2 · Genomics England PanelApp, SMARCA2