IGF2BP2 — the paralogue co-over-expressed with IGF2BP3 in MLL/AF4 B-ALL, and the one with an inhibitor in another leukaemia

{
  "depmap_common_essential": false,
  "chronos_mean": -0.2686,
  "pct_lines_depleted": 4.6,
  "n_lines": 1208,
  "kmt2a_r_lymphoid_diff": -0.1229,
  "kmt2a_r_myeloid_diff": -0.0432,
  "null_percentile": 11.57
}

{
  "(MLL OR KMT2A) AND (IGF2BP2 OR IMP2 OR IMP-2)": 2,
  "IGF2BP2 AND leukemia": 30
}

{
  "chronos_mean_kmt2a_rearranged": -0.3284,
  "chronos_mean_other_leukaemia": -0.2562,
  "difference": -0.0722,
  "pan_essentiality_all_lines": -0.2686,
  "selectivity_rank": 2106,
  "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": "CYTOPLASMIC, WITH THE NUCLEAR VALUE UNSOURCED AND THE GRANULE VALUES SOURCED TO A PAPER THAT NAMES NO GRANULE IN ITS ABSTRACT. UniProt Q9Y6M1 (fetched live 2026-09-21) carries four values: 'Nucleus' and 'Cytoplasm' BOTH WITH EMPTY EVIDENCE LISTS, and 'Cytoplasm, P-body' and 'Cytoplasm, Stress granule' both on one ECO:0000269, PubMed:29476152. The note reads: localised in cytoplasmic mRNP granules containing untranslated mRNAs; localises at the connecting piece and the tail of the spermatozoa; in response to cellular stress such as oxidative stress, recruited to stress granules [E_q93_igf2bp2_loc1].",
  "THE_GRANULE_CITATION_IS_A_RULE_80_CASE_NOT_A_RULE_91_CASE_AND_THE_DISTINCTION_MATTERS": "PubMed:29476152 is Huang et al., 'Recognition of RNA N6-methyladenosine by IGF2BP proteins enhances mRNA stability and translation' (Nat Cell Biol 2018) - which reports that IGF2BPs promote the STABILITY AND STORAGE of target mRNAs under normal AND STRESS conditions, and never names a P-body or a stress granule in its abstract [E_q93_igf2bp2_loc2]. So the compartment measurement is real and sits in a figure, invisible to any title-and-abstract query: IGF2BP2[tiab] AND (\"stress granule\" OR \"P-body\" ...)[tiab] returns 2 records and NEITHER IS 29476152 [E_q93_igf2bp2_loc4]. This is brief rule 80 exactly - the measurement a mechanism graph wants is a supporting panel - and it is why the bounded absence here is written as 'zero in titles and abstracts'.",
  "AND_THE_BETTER_LOCALISATION_PAPER_IS_UNCITED_AND_PARALOGUE_RESOLVED": "Bell and colleagues analysed IGF2BP subcellular localisation and RNP formation by GXXG-to-GEEG conversion of single KH-domain loops in full-length proteins and found that ALL FOUR KH DOMAINS OF IGF2BP1 AND IGF2BP2 PREVENT NUCLEAR ACCUMULATION of these two paralogues AND FACILITATE THEIR RECRUITMENT TO STRESS GRANULES, while the same conversions in IGF2BP3 only modestly affect RNA binding, subcellular localisation and RNA-dependent protein association [E_q93_igf2bp2_loc3]. UniProt does not cite it. IT IS THE PAPER THIS ENTRY MOST NEEDED, because this entry exists to argue that IGF2BP2 and IGF2BP3 may be functionally redundant - and here is a measurement in which they behave DIFFERENTLY, with IGF2BP2 grouping with IGF2BP1 rather than with IGF2BP3.",
  "how_shown": "domain-resolved mutational analysis of subcellular localisation and stress-granule recruitment in full-length paralogues (PMID 23640942); m6A-reader characterisation whose compartment data are in figures rather than the abstract (PMID 29476152)",
  "systems": "human cell lines; none haematopoietic, none leukaemic, none fusion-bearing",
  "node": "localization-igf2bp3-cytoplasmic-rnp-granules-p-bodies-stress-granules"
}

{
  "expressed_in": "UniProt Q9Y6M1: expressed in oocytes, granulosa cells, Leydig cells, spermatogonia and semen at protein level, in testicular cancer at protein level, and WEAKLY in heart, placenta, skeletal muscle, BONE MARROW, colon, kidney, salivary glands, testis and pancreas. Bone marrow is named, and weakly.",
  "in_this_disease": "This entry's existing E1 is the cell-type record and it is B-lineage and human: MLL/AF4-positive B-ALL is characterised by over-expression of BOTH IGF2BP2 and IGF2BP3, while IGF2BP1 marks ETV6/RUNX1 disease and IGF2BP2 is UNDER-expressed in E2A/PBX1. That is expression in the index fusion's own lineage - unusually good for this graph - with no localisation attached.",
  "measured_in_for_localisation": "human cell lines only, and neither localisation paper is haematopoietic."
}

{
  "STRESS_IS_THE_INDUCED_STATE_AND_IT_IS_DOMAIN_ENCODED": "In response to cellular stress such as oxidative stress IGF2BP2 is recruited to STRESS GRANULES [E_q93_igf2bp2_loc1], and that recruitment is FACILITATED BY THE KH DOMAINS - GXXG-to-GEEG conversion of the KH loops impairs it [E_q93_igf2bp2_loc3]. So the induced compartment has a mapped structural requirement, which is rare in this beat.",
  "the_constitutive_state_is_mRNP_STORAGE": "IGF2BP2 resides in cytoplasmic mRNP granules containing UNTRANSLATED mRNAs, and as an m6A reader it promotes the STABILITY AND STORAGE of target transcripts including MYC under normal and stress conditions [E_q93_igf2bp2_loc1][E_q93_igf2bp2_loc2]. Constitutive storage, stress-induced granule - the same architecture as DDX6 in this tranche, reached by a different protein family.",
  "and_MYC_is_the_target_named": "The m6A paper names MYC as an IGF2BP target whose stability is promoted [E_q93_igf2bp2_loc2]. MYC is central to this graph's mechanism entries, so a cytoplasmic, condensate-resident, m6A-dependent stabiliser of MYC mRNA is a post-transcriptional route to a protein this project studies transcriptionally. Nothing of this has been tested in a KMT2A-rearranged cell.",
  "bounded": "No IGF2BP2 localisation, granule or stress 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": "(\"IGF2BP2\"[tiab] OR \"IMP2\"[tiab] OR \"IMP-2\"[tiab] OR \"VICKZ2\"[tiab])",
      "querytranslation": "\"IGF2BP2\"[Title/Abstract] OR \"IMP2\"[Title/Abstract] OR \"IMP-2\"[Title/Abstract] OR \"VICKZ2\"[Title/Abstract]",
      "count": 1388,
      "role": "operand control; IMP2 and IMP-2 are short aliases and inflate it"
    },
    {
      "query": "(gene clause) AND (twelve-term subnuclear clause)",
      "count": 7,
      "title_check": "0 of 7 on topic, and the contamination mechanism is uniform: PMIDs 42446981 (SNORD78 in colorectal cancer), 41387767 (RNA-structure-disrupting SNPs in lincRNAs), 40471366 (SNORD88B and trophoblast), 40044032 (SNORD9 in ovarian cancer), 38545199 (SNHG1 and the miR-7-5p/IGF2BP2 axis), 37565552 (SNHG4 in gastric cancer), 32239639 (SNHG12 in oesophageal cancer). Every one matched through 'nucleolar' inside SMALL NUCLEOLAR RNA or a small-nucleolar-RNA host gene name.",
      "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": "IGF2BP2[tiab] AND (\"stress granule\"[tiab] OR \"stress granules\"[tiab] OR \"P-body\"[tiab] OR \"P-bodies\"[tiab])",
      "querytranslation": "\"IGF2BP2\"[Title/Abstract] AND (\"stress granule\"[Title/Abstract] OR \"stress granules\"[Title/Abstract] OR \"P-body\"[Title/Abstract] OR \"P-bodies\"[Title/Abstract])",
      "count": 2,
      "title_check": "PMIDs 33033522 (an m6A-regulator prognostic signature in endometrial carcinoma, off topic) and 23640942 (the domain-resolved localisation paper cited here). CRITICALLY, UniProt's OWN granule citation - PubMed:29476152 - IS NOT IN THIS SET, because its abstract names no granule. A rule-80 case demonstrated by measurement rather than asserted."
    },
    {
      "query": "(gene clause) AND (nine-term KMT2A/MLL fusion clause)",
      "count": 2,
      "title_check": "PMIDs 25537453 and 21414819; 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."
    }
  ],
  "CONTAMINATED_TOKEN_MEASURED_AT_7_OF_7": "'nucleolar'[tiab] is unusable for any RNA-binding or RNA-adjacent gene. Every one of the seven subnuclear-clause hits for IGF2BP2 came in through SMALL NUCLEOLAR RNA or a small-nucleolar-RNA host gene (SNORD9, SNORD78, SNORD88B, SNHG1, SNHG4, SNHG12). The same mechanism gave 3 of 4 off-topic hits on JAK1 in this tranche. This is the same family as the logged 'speckle*' contamination and it bites hardest on exactly the genes a localisation beat would most want to place in a nucleolus.",
  "reading": "Counted in TITLES AND ABSTRACTS only; supporting figures not searched (brief rule 80), and for this gene the abstract bound is demonstrably the binding constraint, since the compartment citation UniProt itself uses is invisible to the granule query.",
  "absence_class": "measured-not-reported"
}