KHDRBS1 (Sam68) — an RNA-binding protein that holds the MLL-AF4 message DOWN, a class no dropout screen can find

{
  "depmap_common_essential": false,
  "chronos_mean": -0.127,
  "pct_lines_depleted": 4.2,
  "n_lines": 1208,
  "kmt2a_r_lymphoid_diff": 0.087,
  "kmt2a_r_myeloid_diff": 0.059,
  "null_percentile": 91.48
}

{
  "SAM68 AND leukemia": "13 raw, 2 direct functional (both adult T-ALL)",
  "KHDRBS AND leukemia": 1,
  "(MLL OR KMT2A) AND KHDRBS1/Sam68/p68/p62/GSG1": "5 raw, alias-inflated by p62 and p68"
}

{
  "chronos_mean_kmt2a_rearranged": -0.0909,
  "chronos_mean_other_leukaemia": -0.1601,
  "difference": 0.0692,
  "pan_essentiality_all_lines": -0.1273,
  "selectivity_rank": 15942,
  "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."
}

{
  "CONSTITUTIVE": "PREDOMINANTLY NUCLEAR, PARTLY CYTOPLASMIC. UniProt Q07666 (fetched live 2026-09-21) carries 'Nucleus' (ECO:0000269 PubMed:1374686, 26080397, 29496907), 'Cytoplasm' (ECO:0000269 PubMed:29496907) and 'Membrane' (ECO:0000269 PubMed:1374686), with the note 'Predominantly located in the nucleus but also located partially in the cytoplasm'; tissue specificity is ubiquitous with isoform differences in brain, skeletal muscle and liver [E_q93_khdrbs1_loc1].",
  "AND_NONE_OF_THE_THREE_NUCLEAR_CITATIONS_IS_A_LOCALISATION_STUDY_WHILE_THE_ONE_THAT_IS_GOES_UNCITED": "This is the cleanest rule-91 instance in the tranche and it has the same shape as the MEN1 case the standing brief uses as its worked example. (1) PubMed:1374686 is the 1992 cloning of GAP-associated p62 - sequence similarity to a putative hnRNP protein, nucleic-acid binding, arginine dimethylation, SH2-dependent GAP binding - and its abstract describes NO localisation experiment [E_q93_khdrbs1_loc4]. (2) PubMed:26080397 is a SMAR1/HDAC6 alternative-splicing paper in which the protein whose compartment is measured is SMAR1, enriched in nuclear splicing speckles and moved to the cytoplasm by ERK phosphorylation, not Sam68 [E_q93_khdrbs1_loc5]. (3) PubMed:29496907 is a phosphotyrosine-immunoaffinity phosphoproteomics screen for SRMS kinase substrates - 1,258 phosphopeptides over 663 proteins, WITH NO MICROSCOPY AT ALL, and it is cited for BOTH the nucleus and the cytoplasm. MEANWHILE THE FOUNDING SUBNUCLEAR PAPER - PubMed:10473643, titled 'A role for the GSG domain in LOCALIZING Sam68 to novel nuclear structures in cancer cell lines' - IS NOT CITED ON THE RECORD AT ALL [E_q93_khdrbs1_loc2].",
  "how_shown": "the record's own citations are a 1992 cloning paper, a splicing-regulation paper about another protein's compartment, and a phosphoproteomics screen. The real evidence is immunofluorescence plus GFP fusion plus electron microscopy in the uncited 1999 paper, and proteomics plus co-localisation in a 2009 follow-up",
  "systems": "HeLa and other transformed human lines for the SNB work; the 1992 work is recombinant protein from insect Sf9 cells. No leukaemic cell and no KMT2A-rearranged cell",
  "node": "localization-nuclear-matrix-mitotic-scaffold-kmt2a"
}

{
  "expressed_in": "UniProt Q07666: 'Ubiquitously expressed in all tissue examined', with isoform 1 lower in brain, skeletal muscle and liver and isoform 3 intensified in skeletal muscle and liver.",
  "measured_in": "HeLa and other TRANSFORMED human lines for the SNB work [E_q93_khdrbs1_loc2][E_q93_khdrbs1_loc3]; insect Sf9-expressed recombinant protein for the 1992 characterisation; MCF-7 breast cancer cells for the SMAR1/HDAC6 axis.",
  "in_this_disease": "This entry's existing E1 is the only study placing KHDRBS1 on this fusion - MLL-AF4 translation inhibited by ribosomal stalling at AU-rich regulatory sites recognised by KHDRBS-family proteins, with synonymous mutations that disrupt KHDRBS association restoring translation and producing leukaemic transformation. THAT IS A TRANSLATIONAL-CONTROL RESULT AND IT IMPLIES A CYTOPLASMIC SITE OF ACTION, on a protein UniProt calls predominantly nuclear. Nobody has measured Sam68's nuclear-to-cytoplasmic partition, or looked for SNBs, in any KMT2A-rearranged cell.",
  "the_compartment_matters_for_this_entrys_own_mechanism": "Ribosomal stalling happens on cytoplasmic ribosomes. If Sam68 is predominantly nuclear and only partially cytoplasmic, then the fusion-dose brake this entry calls its most important finding is exerted by a minority pool of the protein - which makes the partition, not the total, the quantity that sets fusion dose. No paper measures it."
}

{
  "TRANSFORMATION_IS_THE_DOCUMENTED_STATE_AND_IT_IS_A_POSITIONAL_STATE": "SNBs are predominantly found in TRANSFORMED cells [E_q93_khdrbs1_loc2] and a cell line with low SNB frequency was available to the 2009 study [E_q93_khdrbs1_loc3], so SNB abundance is a variable property of the transformed state rather than a fixed feature of the protein. That is an unusually clean constitutive-versus-induced separation for this beat.",
  "SIGNALLING_IS_THE_SECOND_STATE_AND_IT_RUNS_THROUGH_ACETYLATION": "SMAR1, through HDAC6, maintains Sam68 in a DEACETYLATED state; ERK-1/2 phosphorylation of SMAR1 at T345 and T360 relocates SMAR1 to the CYTOPLASM, preventing its interaction with Sam68 and releasing the inhibitory SMAR1-HDAC6-Sam68 complex, which permits Sam68 acetylation and alternative splicing; loss of heterozygosity at the SMAR1/BANP locus at 16q24.3 in breast cancer cells enhances Sam68 acetylation and CD44 variant exon inclusion [E_q93_khdrbs1_loc5]. So Sam68's ACTIVITY state is set by an upstream protein's COMPARTMENT - an unusual coupling, and the reason UniProt's citation of this paper for Sam68's nucleus is a category slip rather than simple staleness.",
  "and_MITOSIS_is_in_the_name": "Sam68 is Src-Associated in Mitosis, 68 kDa, and the GSG-mutant panel includes a class showing MICROTUBULE association [E_q93_khdrbs1_loc2]. No cell-cycle-resolved localisation of the wild-type protein was located.",
  "bounded": "No KHDRBS1 localisation or cell-state compartment measurement exists in any leukaemic cell, and nothing under a menin inhibitor."
}

{
  "date": "2026-09-21",
  "tool": "NCBI E-utilities esearch, db=pubmed",
  "run_by": "Q93",
  "queries": [
    {
      "query": "(\"KHDRBS1\"[tiab] OR \"Sam68\"[tiab] OR \"SAM68\"[tiab])",
      "querytranslation": "\"KHDRBS1\"[Title/Abstract] OR \"Sam68\"[Title/Abstract] OR \"SAM68\"[Title/Abstract]",
      "count": 503,
      "role": "operand control"
    },
    {
      "query": "KHDRBS1[tiab] AND (\"nuclear bodies\"[tiab] OR \"nuclear body\"[tiab] OR \"perinucleolar\"[tiab])",
      "querytranslation": "\"KHDRBS1\"[Title/Abstract] AND (\"nuclear bodies\"[Title/Abstract] OR \"nuclear body\"[Title/Abstract] OR \"perinucleolar\"[Title/Abstract])",
      "count": 0,
      "role": "THE DECISIVE ALIAS MEASUREMENT, half one - the HGNC symbol finds NOTHING"
    },
    {
      "query": "Sam68[tiab] AND (\"nuclear bodies\"[tiab] OR \"nuclear body\"[tiab])",
      "querytranslation": "\"Sam68\"[Title/Abstract] AND (\"nuclear bodies\"[Title/Abstract] OR \"nuclear body\"[Title/Abstract])",
      "count": 14,
      "role": "THE DECISIVE ALIAS MEASUREMENT, half two - 0 against 14"
    },
    {
      "query": "\"Sam68 nuclear bodies\"[tiab]",
      "querytranslation": "\"Sam68 nuclear bodies\"[Title/Abstract]",
      "count": 8,
      "role": "the compartment has its own NAME and eight papers use it as a phrase"
    },
    {
      "query": "\"SNB\"[tiab] AND Sam68[tiab]",
      "querytranslation": "\"SNB\"[Title/Abstract] AND \"Sam68\"[Title/Abstract]",
      "count": 5,
      "role": "acronym control; SNB is also sentinel node biopsy, so it was paired with the gene name rather than used alone"
    },
    {
      "query": "(gene clause) AND (twelve-term subnuclear clause)",
      "count": 24,
      "bound_not_reproducible_as_stored": "NOT REPRODUCIBLE AS STORED, flagged 2026-09-23. This `query` field records a SUMMARY of the search rather than the search: the clause text it names was never written down, so the string cannot be re-run and this bound could not be included in the cycle-71 audit that re-ran 40 of the corpus's other zero-bounds against the live index (all 40 still returned zero). The clauses have deliberately NOT been reconstructed - guessing them would manufacture a query that was never run, which is worse than an abbreviated one because it would look auditable. This flag does not dispute the count or the absence; it records that this particular bound is not independently checkable."
    },
    {
      "query": "(gene clause) AND (nine-term KMT2A/MLL fusion clause)",
      "count": 2,
      "title_check": "PMIDs 17891136 and 16330219; neither is a localisation measurement.",
      "bound_not_reproducible_as_stored": "NOT REPRODUCIBLE AS STORED, flagged 2026-09-23. This `query` field records a SUMMARY of the search rather than the search: the clause text it names was never written down, so the string cannot be re-run and this bound could not be included in the cycle-71 audit that re-ran 40 of the corpus's other zero-bounds against the live index (all 40 still returned zero). The clauses have deliberately NOT been reconstructed - guessing them would manufacture a query that was never run, which is worse than an abbreviated one because it would look auditable. This flag does not dispute the count or the absence; it records that this particular bound is not independently checkable."
    },
    {
      "query": "(gene clause) AND (localisation clause) AND (fusion clause)",
      "count": 0,
      "bound_not_reproducible_as_stored": "NOT REPRODUCIBLE AS STORED, flagged 2026-09-23. This `query` field records a SUMMARY of the search rather than the search: the clause text it names was never written down, so the string cannot be re-run and this bound could not be included in the cycle-71 audit that re-ran 40 of the corpus's other zero-bounds against the live index (all 40 still returned zero). The clauses have deliberately NOT been reconstructed - guessing them would manufacture a query that was never run, which is worse than an abbreviated one because it would look auditable. This flag does not dispute the count or the absence; it records that this particular bound is not independently checkable."
    }
  ],
  "alias_measurement_verdict": "DECISIVE, 0 against 14, and this is the second such case in one tranche after HCFC2 (0 against 2). The HGNC symbol KHDRBS1 retrieves NOTHING about this protein's named subnuclear compartment; the protein name Sam68 retrieves fourteen records and the compartment's own name, 'Sam68 nuclear bodies', retrieves eight. A localisation beat keyed to current symbols would have written 'no subnuclear compartment assigned' for a protein that has an EPONYMOUS ORGANELLE. Tranche 4 measured that this is a per-gene hypothesis, not a rule, and the running score across six genes now stands at three decisive (SKIC8 2:9:43, HCFC2 0:2, KHDRBS1 0:14), two worthless (SGF29/CCDC101 4 versus 44, KAT2A/GCN5L2/hGCN5 0 versus 0 in fusion context) and one decisive-for-provenance-only (CTR9/p150TSP). The pattern in that is not a rule; the cost asymmetry is.",
  "reading": "Titles and abstracts only; supporting figures not searched (brief rule 80).",
  "absence_class": "never-measured"
}