Oncofusion

start / graph / gene-bicra

Proteinmoderate confidence 3 of 5: One strong study, or several smaller consistent ones.Q47-swi-snf

BICRA (GLTSCR1) - the ncBAF scaffold, zero PubMed records in KMT2A leukaemia, and a lineage-independent dependency in KMT2A-rearranged lines

also called GLTSCR1 · BICRA_HUMAN · glioma tumor suppressor candidate region gene 1

BICRA is the scaffold subunit that, together with BRD9, defines the third and smallest flavour of the SWI/SNF machine. No paper has ever named it in MLL or KMT2A leukaemia, yet in a dependency screen KMT2A-rearranged lines need it in both lineages.

The same thing, for a scientist

BICRA (Q9NZM4, 1,560 aa; primary protein name *BRD4-interacting chromatin-remodeling complex-associated protein*) is the defining scaffold of ncBAF/GBAF, mutually exclusive with its paralogue BICRAL [E1][E2]. The complex it defines contains BRD9, BAF155, a BAF60, SS18, BAF53a and BRG1/BRM and lacks BAF45, BAF47 and BAF57 [E2]; ncBAF localises uniquely to CTCF sites and promoters [E3]. **BICRA has never been named in a KMT2A or MLL paper: `BICRA AND (MLL OR KMT2A)` and `GLTSCR1 AND (MLL OR KMT2A)` both return 0, with operands checked (`BICRA` alone returns 28, `BICRA AND cancer` 12, `GLTSCR1 AND leukemia` 1) [E4].** And yet it ranks 517/17,728 in the KMT2A-rearranged selectivity scan (difference -0.144, pan-essentiality +0.0889 - the average cell line does not need it) and it is one of only four SWI/SNF panel genes depleted in **both** lineage halves (myeloid z -1.14, B-lineage -0.94) [E5][E6]. **This is the most complete gap the beat found: a tractable module with a dependency signature in both lineages and a literature of exactly zero records.**

What it looks like
loading AF-Q9NZM4-F1…
AlphaFold AF-Q9NZM4-F1 ↗
colourshape

This is a prediction, not a measurement. No one has seen this protein in this shape. The colours say how much the software trusts itself, and that is worth reading before anything else — the orange stretches are places the model is effectively saying this part has no fixed shape.

very high (90+)confident (70–90)low (50–70)very low (<50)
I have never read one of these pictures before

A protein is one long chain of amino acids that folds up. The ribbon traces that chain from one end to the other; it is not a wire, it is a simplification of where the backbone runs. Drag to rotate, scroll to zoom.

Domains are the stretches of chain that do a particular job, and they are what the coloured blocks above the picture mark out. Click one and the view flies to it. In this project the important question about any domain is usually whether the fusion still has it — the labels say lost where it does not.

Anything in yellow sticks is a bound small molecule, which is usually the drug.

Go deeper
The three ncBAF genes agree with each other and with nothing else
genecomplex rolegenome-wide rankdifferencepan-essmyeloid zB-lin zsplit verdict
BRD9ncBAF bromodomain39-0.3896+0.0348-1.11-1.62lineage-independent
SS18ncBAF + cBAF224-0.2124+0.0395-0.92-1.27lineage-independent
BICRAncBAF scaffold517-0.1440+0.0889-1.14-0.94lineage-independent
SMARCA4shared ATPase16,333+0.0791-0.3636+0.57-0.23none
ARID1AcBAF-defining12,593+0.0276-0.2337+0.04+0.19none
SMARCB1cBAF/PBAF core14,024+0.0415-0.8470-0.19+0.27none

Three ncBAF genes in the top 3%, all lineage-independent, all with pan-essentiality at or above zero; every cBAF/PBAF-defining and core subunit flat or positive 56. BICRAL, the mutually exclusive scaffold paralogue, does not score in the lineage split (myeloid z +0.27, B-lineage -0.41, verdict weak or ambiguous), so on this evidence the signal is specific to BICRA rather than to the scaffold slot - though two paralogues behaving differently is equally consistent with one of them being noise.

The honest caution is that a near-zero pan-essentiality is exactly what inflates a z-derived rank, and that three subunits of one complex are not three independent observations. But the direction is consistent and the cBAF comparison is the internal control.

Facts and figures
uniprotQ9NZM4
gene symbolBICRA
protein nameBRD4-interacting chromatin-remodeling complex-associated protein
length AA1560
alphafold IDAF-Q9NZM4-F1
accession resolutionUniProt REST gene_exact:BICRA AND organism_id:9606 AND reviewed:true, run 2026-09-20; one reviewed row; gene_primary and length read back from the same record.
complexncBAF (GBAF) scaffold; mutually exclusive with BICRAL
classificationunwritten, unstudied, lineage-independent KMT2A-rearranged dependency in the ncBAF module
subnuclear localizationUNPUBLISHED - a bounded absence, not a NOT APPLICABLE. BICRA is a nuclear chromatin-remodelling scaffold, so a subnuclear compartment is exactly the kind of thing that could exist for it and nobody has looked. Bounded 2026-09-21: (BICRA OR GLTSCR1 OR BRD9 OR ncBAF OR GBAF)[tiab] AND (nucleolar OR "nuclear speckles" OR subnuclear OR condensate OR "nuclear bodies")[tiab] returns ONE record, PMID 39013863, an ARID1A osteoclastogenesis paper that is off-topic at title level - so ZERO on topic, over the whole ncBAF module and not merely over BICRA. Note explicitly: ncBAF's occupancy of CTCF sites and promoters is NOT a subnuclear localisation and is not counted here.
localization shown for fusionNO - nobody has looked, in any KMT2A fusion, in any lineage, and the stronger statement is available here than almost anywhere else in this tranche: BICRA has never been named in a KMT2A or MLL paper at all, under BICRA or under GLTSCR1 (this entry's E4). There is no fusion-context literature in which a localisation could have been buried.
depmap
{
  "rank": 517,
  "of": 17728,
  "percentile": 2.916,
  "difference": -0.144,
  "mean_kmt2a_r": -0.0828,
  "mean_other_leukaemia": 0.0612,
  "pan_essentiality_all_lines": 0.0889,
  "lineage_split_z_myeloid": -1.14,
  "lineage_split_z_b_lineage": -0.94,
  "z_OCI_AML3_NPM1c": -0.02,
  "z_U937_PICALM_MLLT10": -0.07,
  "verdict": "lineage-independent"
}
literature size
{
  "query": "BICRA AND (MLL OR KMT2A)",
  "records": 0,
  "second_query": "GLTSCR1 AND (MLL OR KMT2A)",
  "second_records": 0,
  "operand_controls": {
    "BICRA": 28,
    "BICRA AND cancer": 12,
    "leukemia": 405093
  },
  "complex_form": {
    "query": "ncBAF AND leukemia",
    "records": 15,
    "title_checked": false
  },
  "date": "2026-09-20",
  "tool": "PubMed E-utilities esearch"
}
depmap kmt2a selectivity
{
  "chronos_mean_kmt2a_rearranged": -0.0828,
  "chronos_mean_other_leukaemia": 0.0612,
  "difference": -0.144,
  "pan_essentiality_all_lines": 0.0889,
  "selectivity_rank": 517,
  "rank_denominator": 17728,
  "provenance": "Joined 2026-09-21 from analyses/depmap-kmt2a-selective-scan-full (generated 2026-09-19T10:12:06Z), the project's own genome-wide scan over 11 KMT2A-rearranged and 43 other leukaemia DepMap lines. NOT re-fetched and NOT independent of that scan: every caveat it carries applies here, including that the KMT2A-rearranged arm is a small number of lines drawn from nine individuals, that the rank denominator is 17,728 genes, and that a difference is not a dependency. Added because only 26 of 308 gene entries carried this number, which made any KMT2A-selectivity question answerable for 8% of the corpus.",
  "how_to_read": "A negative difference means MORE required in KMT2A-rearranged lines than in other leukaemia lines. A rank near 17,728 is ANTI-selective. Neither is evidence of a fusion-driven dependency on its own: check pan_essentiality_all_lines before calling anything selective, and note that a paralogue-redundant gene can score near zero for reasons that have nothing to do with this disease - see ana-the-paralogue-redundancy-excuse-is-not-tested-and-my-proxy-decided-the-answer."
}
localization
{
  "subcellular": "NUCLEUS - AND ITS ONLY EXPERIMENTAL CITATION CONTAINS NO LOCALISATION EXPERIMENT. UniProt Q9NZM4 fetched 2026-09-21 carries one SUBCELLULAR LOCATION value, 'Nucleus', on a single attribution: ECO:0000269 PubMed:21555454. That paper was READ IN FULL this cycle (PMC3133372 via the PMC HTML route, 64,734 characters). It contains ZERO occurrences of 'immunofluorescence', 'confocal' or 'localis-', and its three 'localiz-' hits are all about engineering NLS tags onto BRD4 FRAGMENTS so the baits would reach the nucleus. GLTSCR1 appears 43 times and every one is as an interaction partner in Flag-HA affinity proteomics from 293T cells [E_q89_bicra_loc1][E_q89_bicra_loc2].",
  "what_the_nuclear_call_actually_rests_on": "Co-purification from a nuclear complex. BICRA was identified as a BRD4 extraterminal-domain interactor in 293T cells and is a defining scaffold of ncBAF/GBAF, a chromatin-remodelling complex; a chromatin remodeller's scaffold is nuclear by every reasonable inference. The point is that INFERENCE IS WHAT IT IS - there is no image of BICRA anywhere in the record this beat could find, and UniProt's evidence code says otherwise.",
  "subnuclear": "UNPUBLISHED. No subnuclear compartment has been reported for BICRA in any cell type. THE DISTINCTION THAT MATTERS: ncBAF's much-cited unique localisation TO CTCF SITES AND PROMOTERS (this entry's existing E3) is GENOME-WIDE BINDING-SITE OCCUPANCY measured by ChIP, not a subnuclear compartment. A ChIP-seq address and a nuclear compartment are different questions and this entry keeps them apart - the same trap the dispatch brief flagged for CTCF itself, which sits on the other end of that very finding."
}
cell type
{
  "normal": "NOT HAEMATOPOIETIC IN THE ONLY AVAILABLE STATEMENT. UniProt Q9NZM4 tissue specificity reads: 'Expressed at moderate levels in heart, brain, placenta, skeletal muscle, and pancreas, and at lower levels in lung, liver and kidney' [E_q89_bicra_loc1]. BLOOD AND BONE MARROW ARE NOT IN THAT LIST AT ALL - neither as present nor as absent. It is an uncoded bulk-tissue statement of unknown vintage and it simply does not cover the tissue this graph is about.",
  "where_the_biochemistry_was_done": "293T embryonic kidney cells for the BRD4 ET-domain proteomics [E_q89_bicra_loc2]. The ncBAF composition work that defines BICRA's complex was likewise done outside haematopoiesis.",
  "in_this_disease": "THE ONLY CELL-TYPE-LEVEL FACT THIS GRAPH HAS FOR BICRA IS A DEPENDENCY, NOT AN EXPRESSION MEASUREMENT. This entry's existing E5 and E6 record that BICRA ranks 517/17,728 for KMT2A-rearranged selectivity and is depleted in BOTH lineage halves of the KMT2A-rearranged arm (myeloid z -1.14, B-lineage -0.94). A CRISPR dependency in a cell line is evidence that the protein is present and load-bearing in that cell - which is more than the tissue comment gives - but it is not a localisation and not an expression atlas.",
  "grade": "weak. A bulk-tissue list that omits blood, plus a dependency signal. No protein measurement in any haematopoietic cell."
}
cell state
{
  "state": "NONE ESTABLISHED. No BICRA cell-state result exists in any system, let alone a leukaemic one. The gene has ZERO PubMed records in KMT2A/MLL disease under either symbol (this entry's E4), so there is no literature in which a cell state could have been recorded.",
  "the_nearest_thing": "Lineage-independence. BICRA is depleted in both the myeloid and the B-lineage halves of the KMT2A-rearranged arm, which is a statement about which states DO NOT modify the dependency rather than about a state the gene maintains."
}
localization bounded absence
{
  "date": "2026-09-21",
  "tool": "NCBI esearch db=pubmed",
  "queries": [
    {
      "query": "GLTSCR1[tiab] AND nuclear[tiab]",
      "querytranslation": "\"GLTSCR1\"[Title/Abstract] AND \"nuclear\"[Title/Abstract]",
      "count": 1,
      "on_topic": 0,
      "the_one": "33311586 - 'Common variants in SOX-2 and congenital cataract genes contribute to age-related NUCLEAR CATARACT'. The word 'nuclear' is doing entirely different work."
    },
    {
      "query": "(BICRA[tiab] OR GLTSCR1[tiab] OR BRD9[tiab] OR ncBAF[tiab] OR GBAF[tiab]) AND (nucleolar[tiab] OR \"nuclear speckles\"[tiab] OR subnuclear[tiab] OR condensate[tiab] OR \"nuclear bodies\"[tiab])",
      "querytranslation": "(\"BICRA\"[Title/Abstract] OR \"GLTSCR1\"[Title/Abstract] OR \"BRD9\"[Title/Abstract] OR \"ncBAF\"[Title/Abstract] OR \"GBAF\"[Title/Abstract]) AND (\"nucleolar\"[Title/Abstract] OR \"nuclear speckles\"[Title/Abstract] OR \"subnuclear\"[Title/Abstract] OR \"condensate\"[Title/Abstract] OR \"nuclear bodies\"[Title/Abstract])",
      "count": 1,
      "on_topic": 0,
      "the_one": "39013863 - 'ARID1A safeguards the canalization of the cell fate decision during osteoclastogenesis', a different BAF subunit in a different question."
    },
    {
      "query": "BICRA[tiab] OR GLTSCR1[tiab]",
      "querytranslation": "\"BICRA\"[Title/Abstract] OR \"GLTSCR1\"[Title/Abstract]",
      "count": 51,
      "role": "OPERAND CONTROL - the zeros above are zeros inside a real 51-record literature, not zeros from a dead query."
    }
  ],
  "name_list_searched": [
    "BICRA",
    "GLTSCR1 (the retired symbol, run deliberately)",
    "BRD9",
    "ncBAF",
    "GBAF"
  ],
  "complex_form_run": "Yes, per rule 64: ncBAF and GBAF were both searched as complex names alongside the subunit symbols, because a subunit-only enumeration cannot reach a paper that frames everything at the level of the complex. Both returned nothing on topic.",
  "collision_check": "data/symbol-collisions.json was read before searching and carries a BICRA row: the protein name 'BRD4-interacting chromatin-remodeling complex-associated protein' names BRD4, not BRD9, so a reader checking 'is this the ncBAF scaffold?' gets a BET protein. That row also records zero KMT2A/MLL records under either symbol, which this beat did not need to re-derive. The wildcard speckle* was avoided in favour of the quoted phrase, per the speckle* row.",
  "absence_class": "never-measured"
}
How this could bite us back
  • There is no BICRA-directed agent; the module is drugged through BRD9, which means any BICRA-specific contribution is untestable pharmacologically today.
  • If BICRAL can substitute for BICRA, a BICRA-selective agent has an immediate escape route and no assay is watching for it.
  • A pan-essentiality at or above zero is attractive for a therapeutic window and is also the statistic most likely to produce an inflated selectivity rank; the effect size here (-0.14) is small and should be confirmed on raw effect before any programme is built on it.
Still open
  1. Is ncBAF present on KMT2A-fusion-occupied chromatin? A BICRA or BRD9 CUT&RUN in a KMT2A-rearranged line against the fusion's own occupancy has never been published and is the single cheapest decisive experiment in this beat.
  2. Does BICRA loss phenocopy BRD9 degradation in a KMT2A-rearranged line, which would say the dependency is the complex rather than the bromodomain?
  3. Why does BICRAL not score when BICRA does, given they are described as interchangeable scaffolds?
  4. Has anyone ever imaged BICRA? This beat found no immunofluorescence, no fractionation and no tagged-endogenous experiment for the protein in any cell type - only co-purification. If that is right, BICRA is a 1,560-residue scaffold of a named complex with no positional measurement of its own.
  5. ncBAF occupies CTCF sites and promoters by ChIP. Is there a corresponding SUBNUCLEAR organisation - does ncBAF concentrate anywhere visible - or is the CTCF-site preference purely a sequence address with no compartment behind it?
  6. BICRA is depleted in both lineage halves of the KMT2A-rearranged arm and has never been named in a KMT2A or MLL paper. Is the ncBAF module positioned at the same CTCF/loop-anchor addresses this graph implicates in KMT2A breakage, and does that explain the dependency?
Every source behind this entry (8)
  1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    BICRA is UniProt Q9NZM4, 1,560 aa, primary gene BICRA, synonym GLTSCR1; its primary protein name is 'BRD4-interacting chromatin-remodeling complex-associated protein'.

    UniProtKB Q9NZM4 (BICRA_HUMAN) · UniProt Consortium · UniProtKB 2026
    What this evidence cannot show

    gene_exact:BICRA AND organism_id:9606 AND reviewed:true, run 2026-09-20; one reviewed row; gene_primary and length read back from the fetched record. The alternative name 'Glioma tumor suppressor candidate region gene 1' is a positional legacy name and does not assert tumour-suppressor function.

    Evidence that argues against this (1)
    • The primary protein name names BRD4, not BRD9, which is a live trap: a reader checking 'is this the ncBAF scaffold?' gets a string about a BET protein. Only the gene name and the complex literature resolve it. Alpsoy A et al. 2018
  2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

    GLTSCR1 (BICRA) and its paralogue GLTSCR1L (BICRAL) are mutually exclusive components of a third, smaller SWI/SNF subcomplex (GBAF/ncBAF) that also contains BRD9, BAF155, a BAF60, SS18, BAF53a and BRG1/BRM, and lacks BAF45, BAF47 and BAF57.

    PMID 29374058DOI
    What this evidence cannot show

    Defined by density-gradient centrifugation and IP in prostate cancer lines; haematopoietic composition assumed. GBAF displayed ATPase activity and bulk chromatin affinity comparable to BAF despite lacking those subunits.

    Evidence that argues against this (1)
    • GLTSCR1 was dispensable for LNCaP viability while required in PC3 in the same paper, so ncBAF membership does not imply ncBAF dependency and the cell type decides. Xiu Y et al. 2018
  3. E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    ncBAF complexes uniquely localise to CTCF sites and promoters, distinct from cBAF and PBAF addresses.

    A non-canonical SWI/SNF complex is a synthetic lethal target in cancers driven by BAF complex perturbation. · Michel BC, D'Avino AR, Cassel SH, Mashtalir N, et al. · Nat Cell Biol 2018
    PMID 30397315DOI
    What this evidence cannot show

    Mapped in synovial sarcoma and rhabdoid tumour lines, not in leukaemia; the CTCF association is a genome-wide enrichment, not an exclusive localisation.

    Evidence that argues against this (1)
    • If the ncBAF address is CTCF and promoters while KMT2A fusions act through elongation at gene bodies and enhancers, the two are spatially separated and a shared mechanism is not obvious. Shi J et al. 2013
  4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

    BICRA has zero PubMed records in KMT2A-rearranged or MLL leukaemia: `BICRA AND (MLL OR KMT2A)` = 0 and `GLTSCR1 AND (MLL OR KMT2A)` = 0.

    What this evidence cannot show

    Bounded absence. Bounded to PubMed on 2026-09-20 over BICRA and GLTSCR1. Per rule 67 the zero was not trusted until the operands were counted separately and a positive control run through the same code path. Per rule 64 the complex form `ncBAF AND leukemia` was also run and returns 15 records; those were not individually title-checked, so an ncBAF-level result naming no subunit could still exist.

    Evidence that argues against this (1)
    • `ncBAF AND leukemia` returning 15 records means the complex has a leukaemia literature that a BICRA symbol query cannot see - exactly the failure mode rule 64 exists for. The zero is a statement about the symbol, not about the biology. Michel BC et al. 2018
  5. E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    BICRA ranks 517 of 17,728 for KMT2A-rearranged selectivity with difference -0.144 (KMT2A-r -0.0828, other leukaemia +0.0612, pan-essentiality +0.0889 - the average line does not require it).

    Genome-wide KMT2A-rearranged selectivity scan, 17,728 genes (DepMap Chronos via Open Targets) · Oncofusion knowledge graph, analyses/depmap-kmt2a-selective-scan-full/run.py · DepMap Chronos gene effect via Open Targets GraphQL 2026
    What this evidence cannot show

    11 index against 43 comparator lines; a positive pan-essentiality with a small raw difference is precisely the narrowness regime where a z-derived rank flatters a modest effect. The KMT2A-r mean of -0.083 is a weak dependency in absolute terms.

    Evidence that argues against this (1)
    • BICRA's paralogue BICRAL does not score and fails the lineage split, so either the two are not redundant in these cells or the BICRA signal is noise; nothing here distinguishes those. Oncofusion knowledge graph et al. 2026
  6. E6moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

    BICRA is one of four SWI/SNF panel genes depleted in both lineage halves of the KMT2A-rearranged arm (myeloid z -1.14, B-lineage -0.94), alongside BRD9, SS18 and SMARCD1.

    SWI/SNF panel lineage-split DepMap dependency scan, 28 genes · Oncofusion knowledge graph, analyses/swi-snf-lineage-split/run.py · DepMap Chronos gene effect via Open Targets GraphQL 2026
    What this evidence cannot show

    Eight myeloid and five B-lineage index lines, none infant. The two myeloid non-KMT2A reference lines show essentially nothing (OCI-AML3 -0.02, U-937 -0.07), which strengthens the rearrangement association but rests on two lines.

    Evidence that argues against this (1)
    • Three subunits of one complex scoring together is one observation, not three; if ncBAF integrity is what the lines need, BRD9, SS18 and BICRA are measuring the same thing. Alpsoy A et al. 2018
  7. E_q89_bicra_loc1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

    UniProt Q9NZM4 (BICRA_HUMAN, primary gene BICRA, synonym GLTSCR1, 1,560 aa, read back 2026-09-21) carries one subcellular location value, 'Nucleus', on a single attribution ECO:0000269 PubMed:21555454, with no note and no subnuclear qualifier. Its uncoded TISSUE SPECIFICITY comment reads: 'Expressed at moderate levels in heart, brain, placenta, skeletal muscle, and pancreas, and at lower levels in lung, liver and kidney' - a list that does not mention blood, bone marrow or any haematopoietic tissue.

    What this evidence cannot show

    Established as a statement about the database. The tissue list's silence on blood is an ABSENCE FROM A LIST, not a measurement of absence - the comment does not say BICRA is undetectable in haematopoietic tissue, and this entry's own DepMap evidence shows the protein is load-bearing in leukaemia lines, so the list is more plausibly incomplete than contradicted.

    Evidence that argues against this (1)
    • A defender would say the nuclear call needs no imaging because BICRA is a defining scaffold of a chromatin-remodelling complex and chromatin is in the nucleus, so demanding an immunofluorescence is pedantic. That is right about the conclusion and wrong about the record, and the difference bites in a specific way here. This project already learned (rule 91) that 'ECO:0000269' is routinely read as 'a localisation experiment exists', and coverage figures get built on that reading. If BICRA's nuclear call is inference-from-complex-membership, then it belongs in the same evidence class as the by-similarity and rule-based calls this graph downgrades elsewhere - and, more practically, there is then no measurement that could detect BICRA LEAVING the nucleus, or a cytoplasmic pool, or a change under any perturbation. The conclusion is safe; the instrument to test it does not exist. Rahman S et al. 2011
  8. E_q89_bicra_loc2strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

    READ IN FULL. UniProt's sole experimental attribution for BICRA's nuclear localisation describes no localisation experiment on BICRA. The paper is an unbiased Flag-HA affinity-proteomics study of BRD4 fragments in 293T cells that identified NSD3, JMJD6, CHD4, GLTSCR1 and ATAD5 as extraterminal-domain interactors. Over 64,734 characters of stripped full text: 'immunofluorescence' 0 occurrences, 'confocal' 0, 'localis-' 0; 'localiz-' 3, all describing NLS tags fused to BRD4 fragments so the BAITS would reach the nucleus; 'GLTSCR1' 43, all as an interaction partner.

    The Brd4 extraterminal domain confers transcription activation independent of pTEFb by recruiting multiple proteins, including NSD3. · Rahman S, Sowa ME, Ottinger M, Smith JA, Shi Y, Harper JW, Howley PM · Mol Cell Biol 2011
    PMID 21555454DOI
    What this evidence cannot show

    Read in full via the PMC HTML route (PMC3133372) after Europe PMC fullTextXML returned HTTP 500 - the four-route procedure, route 4. The counts are mechanical string counts over stripped text and would miss a method named only inside a figure image or a supplementary file, so the claim is bounded to the article body. The finding is about what the CITATION supports, not about whether BICRA is nuclear; it almost certainly is.

    Evidence that argues against this (1)
    • The honest opposing consideration is that co-purification from a nuclear complex IS positional evidence, and treating it as none would be over-correction. Mashtalir et al. resolved the modular assembly of the SWI/SNF family and placed BICRA as the defining scaffold of the ncBAF/GBAF module, built on ATP-dependent remodelling of nucleosomes - a substrate that exists only in the nucleus. A protein that is a stoichiometric subunit of a nucleosome remodeller is nuclear on grounds far stronger than most immunofluorescence. What the objection cannot rescue is the SPECIFICITY: complex membership constrains the compartment to 'nucleus' and says nothing at all about a subnuclear address, about a non-complexed pool, or about redistribution under perturbation, which are the questions this dimension exists to answer. So the nuclear call stands on better grounds than its citation, and the subnuclear blank is real either way. Mashtalir N et al. 2018
Connected to