SATB1 - a nuclear-matrix and PML-body genome organiser with its own matrix-targeting sequence, the clearest subnuclear-localization entry in this beat, and a T-lineage factor that B cells do not need

No three-dimensional model is available for this protein, so the map above is the whole picture. That is itself informative: AlphaFold does not model chains this long, and the parts that have been crystallised are held under the entries for the individual domains.

• SATB1 promotes HSC quiescence and represses commitment. An agent that reduced SATB1 function would push normal HSCs out of quiescence and into commitment - i.e. exhaust the stem compartment - which is the worst possible on-target toxicity in a child who will need decades of haematopoiesis, and it would not appear in any short-term cytotoxicity assay.
• SATB1 is dispensable for normal B-cell development. That looks like a therapeutic window and is more likely to be a sign that SATB1 is simply not load-bearing in this lineage - the same shape as a target whose knockout phenotype is clean because nothing depends on it. The window and the futility are indistinguishable until someone measures a dependency.
• SATB1 reprograms chromatin organisation to promote growth and metastasis in solid tumours, and SATB1 and SATB2 have opposing roles in at least one reported setting. A SATB1-directed agent in a child therefore carries a paralogue-imbalance risk with a published cancer phenotype attached to the imbalance itself.

1. Does the internal SATB1 promoter at chr3:18,466,065 produce a protein that still carries the NLS (20-40) and the NMTS (224-278)? If not, the species made there is not a nuclear-matrix protein, and 'MLL1-N at SATB1' would mean something quite different. Comparing the four transcripts' coding starts in Ensembl settles it in minutes and was not done.
2. Is KMT2A's own nuclear-matrix association SATB1-dependent? This graph holds localization-nuclear-matrix-mitotic-scaffold-kmt2a, and SATB1 is the protein shown to determine CTCF's matrix retention. Whether it does the same for KMT2A or for a KMT2A fusion has never been asked by anyone.
3. Is SATB1 a dependency in KMT2A-rearranged leukaemia despite being dispensable for normal B-cell development? A developmental null does not exclude a leukaemic requirement, and the four-way literature zero means the experiment has never been run in any KMT2A system.
4. SATB1 level partitions HSCs into interconvertible lymphoid-biased and myeloid-biased states. KMT2A-rearranged leukaemia has a live cell-of-origin question and a documented lineage-switch phenomenon. Does SATB1 level track lineage switch at relapse in KMT2A-rearranged disease? This is answerable from existing paired diagnosis-relapse expression data.
5. The 2003 SATB1 'cage-like network' is a microscopy description from before the condensate era. Does the SATB1 network satisfy this graph's own localization-condensate-evidence-standards, and if not, what is it made of?

1. E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   SATB1 is Q01826 (SATB1_HUMAN), 763 aa, with a CMP domain at 71-172, a CUTL domain at 175-248, CUT1 and CUT2 DNA-binding domains at 361-448 and 484-571, a homeobox DNA-binding domain at 645-704, a nuclear localisation signal at 20-40, a protein-interaction motif at 139-143 and a NUCLEAR MATRIX TARGETING SEQUENCE at 224-278; it is localised to the nucleus matrix and to PML bodies, is expressed predominantly in thymus, binds the consensus SATB1-binding sequence at nuclear matrix- or scaffold-associated regions, acts as a docking site for chromatin-remodelling enzymes and recruits HDACs or HATs directly to promoters and enhancers.

   UniProtKB Q01826 (SATB1_HUMAN), DNA-binding protein SATB1 · UniProt Consortium · UniProt 2026
   ▸What this evidence cannot show

   Fetched live 2026-09-23 via gene_exact:SATB1 AND organism_id:9606 AND reviewed:true, which returned exactly ONE reviewed record, so rule 62 does not fire. data/symbol-collisions.json (208 rows) word-boundary grepped for SATB1 before the first query: no row, and HGNC lists neither a previous symbol nor any alias symbol, so rule 69's retired-symbol arm has nothing to run against. Per rule 91 the papers UniProt cites for the subcellular comment were not opened, and the comment is not resolved to an isoform - which matters here because an alternative internal promoter (E9) could produce a species lacking the NLS at 20-40 or the NMTS at 224-278. 'Expressed predominantly in thymus' is a tissue summary and is contradicted in scope by the HSC and B-cell literature cited below.

   ▸Evidence that argues against this (1)

   • A subcellular-location comment of 'Nucleus matrix' is exactly the kind of annotation this project has found to be under-sourced elsewhere: rule 91 records that UniProt's coded experimental citations for MEN1, KMT2A, AFF1 and AFF4 localisation each point at a paper that is not the founding or best observation. Here the founding observation IS available and is cited directly as E3, so this entry does not rest on the database row. Nat Genet 2003
2. E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database

   HGNC assigns SATB1 (HGNC:10541, 'SATB homeobox 1') the primary Ensembl gene id ENSG00000182568, cytoband 3p24.3 and UniProt Q01826, with NO previous symbol and NO alias symbol; Ensembl places the gene at GRCh37 chr3:18,386,879-18,487,080 and GRCh38 chr3:18,345,056-18,445,621, both on the MINUS strand.

   HGNC record HGNC:10541 (SATB1) and Ensembl gene ENSG00000182568 on GRCh37 and GRCh38 · HUGO Gene Nomenclature Committee; Ensembl · genenames.org / Ensembl 2026
   ▸What this evidence cannot show

   Fetched live 2026-09-23 and resolved before any sentence using them was written. An empty alias list means the symbol-based literature bound in E10 is not undercounting by an alias arm - the rarest clean case in this beat. The MINUS strand is load-bearing for reading the Q118 bin (E9). Note that SATB2 is a paralogue with opposing roles in at least one reported setting, so a family-level search is not a gene-level search even though the symbol itself is clean.

   ▸Evidence that argues against this (1)

   • A completely empty alias and previous-symbol record is unusual enough to be worth doubting rather than celebrating. It means HGNC has never had to reconcile a competing name for this gene - but the protein is referred to throughout the older literature as 'the MAR-binding protein' and 'special AT-rich sequence binding protein 1' in full, neither of which is a symbol and neither of which a symbol-based [tiab] query reaches. The clean HGNC row does not make the literature bound in E10 complete. Genes Dev 2000
3. E3established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

   In thymocyte nuclei SATB1 has a cage-like 'network' distribution circumscribing heterochromatin and selectively tethers specialised DNA sequences onto that network, shown by FISH on wild-type and Satb1-null thymocytes using in-vivo SATB1-bound sequences as probes; many gene loci including Myc are anchored by the SATB1 network at specific genomic sites and the anchoring correlates precisely with correct regulation of distant genes; across a 70 kb gene-enriched region, acetylation of histone H3 at Lys9 and Lys14 peaks at the SATB1-binding site and extends over roughly 10 kb covering SATB1-regulated genes, whereas in Satb1-null thymocytes the same site is marked by H3 Lys9 methylation.

   Tissue-specific nuclear architecture and gene expression regulated by SATB1 · Cai S, Han HJ, Kohwi-Shigematsu T · Nat Genet 2003
   PMID 12692553DOI
   ▸What this evidence cannot show

   Abstract-level read; the full text was not opened this session. MOUSE THYMOCYTES. The acetylation-to-methylation switch is measured across one 70 kb region, not genome-wide, and 2003-era histone-modification analysis is lower-resolution than a modern ChIP-seq. The 'cage-like network' is a microscopy description; this entry reports it as such and does not import a biophysical model of what the network is made of.

   ▸Evidence that argues against this (1)

   • A microscopy-defined 'network' circumscribing heterochromatin is the class of observation that phase-separation-era work has repeatedly reinterpreted, and this graph already holds localization-condensate-evidence-standards for exactly that reason. Whether the SATB1 network is a proteinaceous matrix, a condensate, or an imaging artefact of fixation is not settled by the 2003 experiment and is not settled here. Iqbal Lab oncofusion knowledge graph 2026
4. E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   SATB1 colocalises extensively with CTCF throughout the genome; depletion of SATB1 in human and murine T cells causes transcriptional dysregulation of T-cell-activation genes and alters 3D genome architecture at the level of compartments, topologically associating domains and loops, increasing chromatin contacts among and across SATB1/CTCF co-occupied sites; loss of SATB1 does not affect CTCF occupancy but significantly reduces the retention of CTCF in the nuclear matrix.

   SATB1 regulates 3D genome architecture in T cells by constraining chromatin interactions surrounding CTCF-binding sites · Wang B, Ji L, Bian Q · Cell Rep 2023
   PMID 37000624DOI
   ▸What this evidence cannot show

   Abstract-level read; the full text was not opened this session. HUMAN AND MURINE T CELLS - this is a T-lineage 3D-genome result and there is no equivalent measurement in a B-lineage or leukaemic cell. 'Reduces retention of CTCF in the nuclear matrix' is a fractionation readout, and fractionation-defined matrix association is an operational definition that depends on extraction conditions.

   ▸Evidence that argues against this (1)

   • SATB1 has been 'long proposed to act as a global chromatin loop organizer in T cells', and this paper's own framing is that its exact functions remained elusive - i.e. the field's prior model of SATB1 as a loop-former was revised into SATB1 as a loop-CONSTRAINER, since its loss INCREASES contacts. A long-standing structural model for this protein has already been inverted once by better data. Cell Rep 2023
5. E5established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper

   SATB1-null mice are small, have disproportionately small thymi and spleens and die at three weeks; triple-negative thymocytes are greatly reduced and thymocyte development is blocked mainly at the double-positive stage; the few peripheral CD4+ single-positive cells undergo apoptosis and fail to proliferate on activation; and at least 2% of 589 genes examined, including a proto-oncogene, cytokine receptor genes and apoptosis-related genes, are derepressed at inappropriate stages, with Il2ra and Il7r ectopically transcribed in double-positive thymocytes.

   The MAR-binding protein SATB1 orchestrates temporal and spatial expression of multiple genes during T-cell development · Alvarez JD, Yasui DH, Niida H, Joh T, Loh DY, Kohwi-Shigematsu T · Genes Dev 2000
   PMID 10716941
   ▸What this evidence cannot show

   Abstract-level read; the full text was not opened this session, and no DOI is recorded in this record's PubMed ArticleIdList. MOUSE germline knockout, T-LINEAGE. The '2% of 589 genes' figure is from a 2000-era array with a small gene panel and is a floor, not a genome-wide estimate.

   ▸Evidence that argues against this (1)

   • The same germline model that produces a striking T-cell phenotype leaves B-cell development essentially untouched when SATB1 is deleted from the pro-B stage, so the dramatic null phenotype is lineage-restricted and must not be read as a general haematopoietic requirement. Cell Mol Immunol 2023
6. E6strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   Satb1 is a critical regulator of haematopoietic stem cell fate: HSCs lacking Satb1 have defective self-renewal, are less quiescent and show accelerated lineage commitment through less symmetric self-renewal and more symmetric differentiation divisions, with Satb1 simultaneously repressing genes for HSC activation and cellular polarity including Numb and Myc. Separately, in Satb1/Tomato-knockin reporter mice, SATB1-high HSCs differentiate preferentially toward the lymphocytic lineage while SATB1-negative HSCs follow the myeloid lineage, and the two populations are interconvertible upon transplantation.

   Satb1 regulates the self-renewal of hematopoietic stem cells by promoting quiescence and repressing differentiation commitment · Will B, Vogler TO, Bartholdy B, Garrett-Bakelman F, Mayer J, Barreyro L, et al. · Nat Immunol 2013
   PMID 23563689DOI
   ▸What this evidence cannot show

   Abstract-level read; the full text was not opened this session. MOUSE. The reporter-based lineage-bias half of this claim comes from a SEPARATE paper (PMID 29898394, Cell Rep 2018, Doi et al., conditional Satb1 knockout and Satb1/Tomato knock-in) and is attributed here in the claim text rather than merged into the 2013 citation. A third paper (PMID 23791645, Immunity 2013) reports Satb1 directing HSC differentiation toward lymphoid lineages and was seen at TITLE LEVEL ONLY; it is not relied upon.

   ▸Evidence that argues against this (1)

   • Satb1 promoting quiescence and repressing commitment, and SATB1-high HSCs being lymphoid-biased, are two different claims from two groups, and the interconvertibility of the SATB1-high and SATB1-low populations on transplantation means SATB1 level is a state variable rather than a fixed identity. A leukaemic cell's SATB1 level therefore says nothing durable about its cell of origin. Doi Y et al. 2018
7. E7strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper

   In a mouse model in which SATB1 is deleted from the pro-B-cell stage, SATB1 is DISPENSABLE for B-cell development and for establishing a broad immunoglobulin heavy-chain repertoire - unlike its essential role in T cells - and in mature B cells it has an ambivalent function, acting sequentially as a positive and then a negative regulator of immunoglobulin gene transcription in naive and activated cells, with its negative regulatory function extending to the germinal-centre response where it limits somatic hypermutation.

   A dual function for the chromatin organizer Special A-T rich Binding Protein 1 in B-lineage cells · Thomas M, Bruzeau C, Martin OA, Pollet J, Bender S, Carrion C, et al. · Cell Mol Immunol 2023
   PMID 37544964DOI
   ▸What this evidence cannot show

   Abstract-level read; the full text was not opened this session. MOUSE conditional knockout from the pro-B stage. 'Dispensable for B-cell development' is a normal-development result and says nothing about whether SATB1 is required in a TRANSFORMED B-lineage cell - a leukaemic dependency can exist where a developmental requirement does not, and this graph holds several examples. This citation is included precisely because it is the strongest argument AGAINST SATB1 mattering in this disease.

   ▸Evidence that argues against this (1)

   • A gene dispensable for normal development of a lineage can still be a dependency in that lineage's leukaemia - the entire premise of a synthetic or state-specific vulnerability. The correct reading of this paper is that it removes the developmental prior, not that it settles the leukaemic question, and nobody has asked the leukaemic question: SATB1[tiab] AND KMT2A[tiab] returns 0. NCBI PubMed 2026
8. E8moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset

   In RS4;11 (KMT2A::AFF1) the hg19 2 kb bin chr3:18,466,000 reaches 73.4x the MLL1-N track's own genome median and is in the top thirty bins genome-wide; under VTP-50469 its MLL1-N within-sample share ratio is 0.618 and its MENIN share ratio is 0.150, the highest MENIN ratio among the top thirty, whose range is 0.030 to 0.296.

   Q118 re-analysis of GSE127507, instrument B: 2 kb genome-wide bins from the deposited .deduped.tdf files (RS4;11 MLL1n GSM3635695/96, MENIN GSM3635693/94, input GSM3635699); analyses/q118-menin-at-the-human-index-fusion/results-instrumentB.json, run 2026-09-23 · Iqbal Lab oncofusion knowledge graph 2026
   ▸What this evidence cannot show

   THE MEASUREMENT THAT NAMED THIS GENE IS NOT FUSION-SPECIFIC. The track is MLL1n, Bethyl A300-086A, an MLL1 N-TERMINUS antibody whose epitope is retained in KMT2A::AFF1, so it reads fusion and wild-type MLL1 together at a masking factor the same beat measured as 3.54-4.00x. n = 1 per antibody per condition - GSE127507 has no replicate - and the depositors measured a Drosophila spike-in and deliberately did not apply it, so cross-condition figures are within-sample shares. 'Highest MENIN ratio in the top thirty' is a RANK WITHIN THIRTY, not a statistic: the values 0.150 and 0.124 are not distinguishable at n = 1, no interval can be put on either, and per-bin share ratios at single 2 kb bins are the noisiest quantity in the source beat. It is reported as an observation worth a sentence and explicitly not as a finding.

   ▸Evidence that argues against this (1)

   • A bin where menin leaves LESS completely and MLL1-N persists MORE is exactly what a predominantly wild-type MLL1 site should look like in an antibody that cannot tell the two apart. The conservative reading of SATB1's numbers is that this bin is the least fusion-like in the top thirty, not the most interesting. Iqbal Lab oncofusion knowledge graph 2026
9. E9strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database

   SATB1 is on the MINUS strand and the bin centre chr3:18,466,000 sits at an internal transcription-start cluster: ENST00000417717 starts at GRCh37 18,466,065 (65 bp), ENST00000457005 at 18,466,232, ENST00000338745 at 18,466,829 and ENST00000450898 at 18,466,861 - four protein-coding transcripts within 900 bp - while the gene's 5'-most annotated start on the minus strand is at 18,487,080, 21,080 bp away.

   Ensembl GRCh37 REST: lookup/id/ENSG00000182568?expand=1, per-transcript start coordinates and strand · Ensembl · Ensembl GRCh37 archive 2026
   ▸What this evidence cannot show

   Fetched live 2026-09-23, resolved from a stable gene id to transcript coordinates rather than by name lookup. Ensembl gene models, not experimentally mapped start sites in a leukaemic cell. Which of the four transcripts is used in RS4;11 is NOT determined here, and the inference that an alternative first exon might omit the NLS (20-40) or the NMTS (224-278) is stated as a question, not as a result - the transcripts' coding starts were not compared and doing so would settle it.

   ▸Evidence that argues against this (1)

   • A cluster of four annotated starts within 900 bp is also what an annotation pile-up at a single real promoter looks like, and a 2 kb bin cannot resolve them. Calling this an 'internal alternative promoter' rather than 'the promoter' rests on the 21 kb gap to the 5'-most start, which is a gene-model statement and not a measurement of transcription in this cell. Ensembl GRCh37 archive 2026
10. E10moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database

    Bounded literature counts, 2026-09-23: SATB1[tiab] 666; SATB1[tiab] AND KMT2A[tiab] 0; SATB1[tiab] AND MLL[tiab] 0; SATB1[tiab] AND "MLL-AF9"[tiab] 0; SATB1[tiab] AND "MLL-AF4"[tiab] 0; SATB1[tiab] AND leukemia[tiab] 22; SATB1[tiab] AND "nuclear matrix"[tiab] 61; SATB1[tiab] AND "hematopoietic stem"[tiab] 11; SATB1[tiab] AND "B cell"[tiab] AND development[tiab] 8. HGNC records no alias and no previous symbol for SATB1, so there is no second symbol arm to run.

    PubMed esearch via scripts/pubmed_safe.py, 2026-09-23 · Q119-genes-the-measurement-named · NCBI PubMed 2026
    ▸What this evidence cannot show

    Bounded by TITLES AND ABSTRACTS only (rule 80). One tag per term, no parenthesised group (rule 94); every querytranslation was read back and each expanded to "TERM"[Title/Abstract], and each zero carried a non-empty translation with the 'No items found.' warning, i.e. they are genuine zeros by the pubmed_safe signature and not instrument failures. This is the STRONGEST absence in the beat: SATB1 has a 666-record literature and a 22-record leukaemia literature and ZERO records pairing it with KMT2A or MLL under any spelling tried. The leukaemia records were read at title level and are predominantly T-ALL, AML and lymphoma; none is a KMT2A-fusion paper.

    ▸Evidence that argues against this (1)

    • A four-way zero across KMT2A, MLL, MLL-AF9 and MLL-AF4 is a strong bibliographic absence and a weak biological one: SATB1 is a genome organiser whose occupancy would appear in a supplementary Hi-C or ChIP table of a KMT2A paper without ever entering a title or abstract, which is precisely the failure rule 80 was written from. The absence is real about the indexed literature and unproven about the archives. Cancer Gene Ther 2026

• related to →Nuclear matrix and mitotic chromosome scaffold — KMT2A sits with topoisomerase II, in the same place the gene breaks

  SATB1 carries its own nuclear matrix targeting sequence (224-278) and determines CTCF's matrix retention; whether it does the same for KMT2A has never been asked

• regulates →CTCF - named in the id of an existing entry in this graph and having no entry of its own, and the protein whose binding sites break preferentially in KMT2A-rearranged patients

  colocalises extensively with CTCF; SATB1 loss does not change CTCF occupancy but significantly reduces CTCF retention in the nuclear matrix and increases contacts across co-occupied sites

• regulates →MYC - named in a hundred entries of this graph and, until now, written in none

  the MYC locus is anchored onto the SATB1 nuclear network at specific genomic sites; Satb1-null HSCs derepress Myc and Numb

• regulates →Quiescence (G0) — the state 15 entries invoke to explain measurable residual disease, defined here for the first time

  Satb1 promotes HSC quiescence and represses differentiation commitment; Satb1-null HSCs are less quiescent and commit faster

• expressed in →Bone marrow haematopoietic stem cell — the postnatal comparator, and why the same fusion behaves differently here

  SATB1 level partitions HSCs into interconvertible lymphoid-biased (SATB1-high) and myeloid-biased (SATB1-low) populations

• regulates →H3K9ac - the mark KMT2A fusions pile up on their own targets, the one the AF9 reader actually binds, and the one a DOT1L inhibitor strips off first

  H3K9/K14 acetylation peaks at the SATB1-binding site and spreads ~10 kb; in Satb1-null thymocytes the same site carries H3K9 methylation instead

• supports →Instrument check: the top thirty RS4;11 MLL1-N 2 kb bins recovered from a deposited .tdf are JMJD1C (259x the genome median), ZMYND11, BAHCC1, CDK6, MYB, RUNX2, IGF2BP2, FOXP1, SATB1 and ARID1B - canonical KMT2A::AFF1 targets, annotated only against the series' own RefGene windows

  the entry that named this gene; resolved here to an internal SATB1 transcription-start cluster 21 kb from the canonical start

• related to →Pro-B cell — the stage at which KMT2A::AFF1 leukaemia is arrested, and the phenotype the diagnosis is made on

  SATB1 deleted from the pro-B stage leaves B-cell development and the IgH repertoire intact - the published argument AGAINST SATB1 mattering in this lineage