Every upstream NF-kB component this beat enumerated, with its query, its count, and why it did or did not get an entry
also called NF-kB upstream enumeration · Q38 bounded absences · the IKK beat's rejection list
This is the working record of which NF-kB proteins were checked, what each search returned, and which ones were dropped. Writing down what was looked at and rejected is as useful as writing down what was found, because otherwise the next person cannot tell a gap from a decision.
The same thing, for a scientist
**Eighteen symbols were enumerated, all resolved to UniProt by `gene_exact` with `gene_primary` requested and every row read.** Nine became entries; nine are recorded here as bounded absences or bounded rejections. **The live trap, caught this session:** `gene_exact:TRAF2 AND organism_id:9606 AND reviewed:true` returns **two** reviewed rows and the first is **Q92844 TANK_HUMAN**, which carries TRAF2 as a synonym; the intended record is **Q12933 TRAF2_HUMAN** (501 aa). That is Rule 62 reproduced live, and no other symbol in this enumeration had the problem. **The load-bearing counted absences.** `IKBKG`/NEMO: **0**. `c-Rel`: **0**. `(KMT2A OR MLL) AND infant[tiab] AND NF-kB`: **0**. `menin/revumenib/ziftomenib AND IKK/NF-kB inhibitor/bortezomib`: 3 records, **0 after title-checking** — no published combination of a menin inhibitor with any NF-kB-directed agent exists in any system. **The methodological finding this beat would want passed on.** The 2013 Cancer Cell paper knocks down **Ikka, Ikkb, Ikkg and IRAK3** and is **unreachable from any of those four symbols**, because its abstract frames at the level of 'IKK/NF-kB'. A subunit-keyed absence count for any component of a named complex will therefore be wrong in the same direction every time. This is the NSL/Mediator lesson in the brief, reproduced for IKK, and it means the absences below bound **framed literature**, not **measurement**.
The enumeration table
| symbol | UniProt (primary read back) | aa | KMT2A/MLL records | title-checked on topic | outcome |
|---|---|---|---|---|---|
| IKBKB | O14920 | 756 | 5 | 3 | entry gene-ikbkb |
| CHUK | O15111 | 745 | 4 | 0 | entry gene-chuk (evidence is inside papers, not findable by symbol) |
| IKBKG | Q9Y6K9 | 419 | 0 | 0 | entry gene-ikbkg (absence is the content) |
| NFKBIA | P25963 | 317 | 2 | 1 (non-leukaemic) | entry gene-nfkbia |
| RELB | Q01201 | 579 | 4 | 3 | entry gene-relb |
| REL (c-Rel) | Q04864 | 619 | 0 | 0 | entry gene-rel (one null measurement) |
| MAP3K14 (NIK) | Q99558 | 947 | 3 | 1 | entry gene-map3k14 |
| TNF | P01375 | 233 | 37 | several, incl. KMT2A::AFF1 | entry gene-tnf |
| TIFAB | Q6ZNK6 | 161 | 3 | 2 (both KMT2A fusions) | entry gene-tifab |
| MAP3K7 (TAK1) | O43318 | 606 | 3 | 0 | bounded absence |
| TRAF6 | Q9Y4K3 | 522 | 3 | 0 TRAF6-focused | bounded rejection |
| TRAF2 | Q12933 (not Q92844 TANK) | 501 | 2 | 0 | bounded absence + accession trap |
| MYD88 | Q99836 | 296 | 7 | 0 | bounded absence |
| IRAK1 / IRAK4 | P51617 / Q9NWZ3 | 712 / 460 | 5 combined | 2 (AML-wide) | folded into mechanism-il1-irak-axis-in-leukaemia-stem-cells |
| IRAK3 | Q9Y616 | 596 | — | screen hit only | folded into the same mechanism |
| TNFRSF1A | P19438 | 455 | 2 | 0 | bounded absence |
| IL1R1 | P14778 | 569 | 25 raw | 1 (AML-wide) | folded into the IL-1/IRAK mechanism |
| NFKBIB / NFKBIE | Q15653 / O00221 | 356 / 500 | 2 | 0 | bounded absence |
| TNFAIP3 (A20) | P21580 | 790 | not separately searched | — | named only as a NIK-induced feedback gene; no entry |
Bounded rejections — things a reader would expect cited, and why they are not
TRAF6 (PMID 36453131). IMPDH inhibition overactivates TLR-TRAF6-NF-kB and VCAM1 and suppresses MLL-fusion AML. On topic and important, but it is not a TRAF6 dossier: TRAF6 is a named waypoint in a pathway the paper drives rather than a component it manipulates. It is cited in mechanism-il1-irak-axis-in-leukaemia-stem-cells as a direction-challenging result, which is what it is. What would change this: a TRAF6 loss-of-function experiment in a KMT2A-rearranged model.
LAMP5 (PMID 33910331). A direct MLL-fusion target that routes innate-immune signalling toward NF-kB and away from type-I interferon, and whose depletion inhibits NF-kB signalling. Squarely on beat — and already in this graph as cell-state-innate-immune-lamp5. Not duplicated; linked.
TRAF2 DepMap signal. TRAF2 came back as a weak DepMap hit in an unrelated triage in this project and failed a permutation null at p = 0.1618 (ana-the-null-kills-my-own-shortlist). The dependency number is not cited anywhere in this folder as support for anything. What would change this: a KMT2A-specific TRAF2 experiment, of which there are none.
PLK1 and PPP4C. Both are screen hits alongside the IKK subunits, and PLK1 is both an NF-kB target and an MLL-AF9-regulated gene 1. Dropped because PLK1 is a mitotic kinase with a large independent literature and folding it into an NF-kB dossier would misrepresent why it scores. What would change this: evidence that PLK1 inhibition changes fusion occupancy rather than killing dividing cells.
Bortezomib as an 'NF-kB inhibitor'. Considered and deliberately separated into mechanism-proteasome-inhibition-splits-by-lineage-in-mll-leukaemia, because in the index lineage its documented mechanism has no NF-kB in it.
Amlexanox. Used in KMT2A::AFF1 B-ALL as an S100A6 downregulator and separately a known TBK1/IKK-epsilon inhibitor. Recorded in gene-tnf with the explicit warning that the two activities have never been connected, rather than claimed as an IKK-family result.
Verteporfin. Arrived by Connectivity Map signature matching on NIK-stabilised LSCs, not by acting on NIK. Recorded in gene-map3k14 with that distinction stated.
Search hygiene notes for whoever re-runs this
- Every
querytranslationwas read. No[UID]artefact appeared in any zero. t(4;11)was written without any other parenthesis in its query, per the brief's rule; PubMed rendered it as the phraset 4 11.- Short, word-like aliases were deliberately excluded: bare
REL, bareNIK,I-Rel. Including them inflates counts with noise, which is the RD/NELFE failure the brief records. The counts above are therefore conservative in recall and honest in precision. - Aliases not searched, and which could add records:
IKK1for CHUK,IKK2for IKBKB (partly covered by PubMed's own expansion),FIP3/IKKAP1for IKBKG,MAD-3for NFKBIA,TGF-beta activated kinase 1as a phrase for MAP3K7. - Search the complex's name as well as its subunits.
IKKandNF-kBas phrases reach the founding paper; no subunit symbol does.
| script | none - manual enumeration via UniProt REST and PubMed E-utilities, scratchpad files prefixed q38_ |
|---|---|
| status | run |
| result summary | 18 upstream NF-kB symbols enumerated; 9 became entries, 9 recorded as bounded absences or rejections. IKBKG, c-Rel, infant+NF-kB and menin+NF-kB-agent all return 0 after title-checking. The TRAF2/TANK gene_exact trap reproduced live. The founding paper measures four enumerated genes and is unreachable from any of their symbols. |
inputs
[ "UniProt gene_exact queries with gene_primary read back, 22 symbols", "PubMed esearch/esummary with querytranslation inspected, 23 queries", "Europe PMC and efetch db=pmc full text for PMC3816582 and PMC5810947" ]
entries written
[ "gene-ikbkb", "gene-chuk", "gene-ikbkg", "protein-ikk-complex", "gene-nfkbia", "gene-relb", "gene-rel", "gene-map3k14", "gene-tnf", "gene-tifab", "mechanism-ikk-nfkb-retains-mll-on-chromatin", "mechanism-canonical-versus-non-canonical-nfkb-in-kmt2a-rearranged-leukaemia", "mechanism-nfkb-lineage-split-myeloid-story-lymphoid-silence", "mechanism-proteasome-inhibition-splits-by-lineage-in-mll-leukaemia", "mechanism-il1-irak-axis-in-leukaemia-stem-cells", "hyp-ikk-inhibition-is-a-menin-independent-chromatin-eviction-route", "exp-ikk-plus-menin-inhibitor-occupancy-combination", "risk-ikk-inhibition-essential-immune-function" ]
- A bounded absence written from a subunit symbol is systematically wrong for any protein that publishes at complex level, and this beat found the founding paper of its own field to be exactly that case. Absences in this graph should be read with that error bar.
- Excluding short aliases for precision means these counts under-report. A future reader should not treat a zero here as stronger than it is.
- How many other complex-level beats in this graph carry subunit-keyed absence counts that are wrong in the same direction? The IKK case gives a way to test: find any entry claiming zero records for a subunit of a named complex and re-run the search on the complex's name.
- Do the unsearched retired and case-variant aliases (IKK1, FIP3, MAD-3, 'TGF-beta activated kinase 1') change any of the zeros above?
- Is there a KMT2A-specific TRAF6 or TAK1 experiment in a preprint or a supplementary figure that a title/abstract search cannot reach?
- Would a full-text search engine rather than title/abstract change the shape of these counts? Every absence here is an abstract-level absence.
- E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
The functional genomic screen identified 33 candidate kinases or phosphatases whose knockdown impaired growth of MLL-oncogene-transformed myeloid progenitors by 50% or more, among them the IKK complex (IKKa, IKKb, IKKg), polo-like kinase 1 (PLK1), protein phosphatase 4 catalytic subunit (PPP4C) and IL-1R-associated kinase 3 (IRAK3). PLK1 is downstream, since RELA transcriptionally activates its promoter through direct binding, and PLK1 is also an MLL-AF9-regulated gene. Despite knocking down Ikka, Ikkb, Ikkg and IRAK3, the paper's title and abstract name none of these symbols, framing instead at the level of 'IKK/NF-kB'.
Epigenetic roles of MLL oncoproteins are dependent on NF-κB. · Kuo HP, Wang Z, Lee DF, Iwasaki M, Duque-Afonso J, Wong SHK, Lin CH, Figueroa ME, Su J, Lemischka IR, Cleary ML · Cancer Cell 2013What this evidence cannot show
Full text read this session via efetch db=pmc (PMC3816582) after Europe PMC fullTextXML returned HTTP 500 - the documented route-2-fails-route-3-succeeds pattern. The claim that the abstract names no subunit was verified against the fetched abstract, not assumed.
Evidence that argues against this (1)
- A screen that scores growth impairment on knockdown identifies what is required in the model used, and this one was run in MLL-oncogene-transformed MYELOID cells. The index disease here is infant B-lineage KMT2A::AFF1, and the same pathway's two arms behave oppositely in myeloid leukaemia - canonical signalling sustains leukaemia stem cells while stabilising the non-canonical kinase NIK suppresses MLL-AF9 disease - so a kinase list from one arm in one lineage does not transfer. Xiu Y et al. 2018
- E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
UniProt accession resolution trap reproduced live on 2026-09-20: gene_exact:TRAF2 AND organism_id:9606 AND reviewed:true returns two reviewed human rows. The first is Q92844 (TANK_HUMAN), primary gene name TANK, 425 aa, which carries TRAF2 as a synonym. The intended record is Q12933 (TRAF2_HUMAN), primary gene name TRAF2, 501 aa. Every other symbol in this enumeration - CHUK, IKBKB, IKBKG, NFKBIA, NFKBIB, NFKBIE, RELB, REL, MAP3K14, MAP3K7, TRAF6, MYD88, IRAK1, IRAK4, IRAK3, TNF, TNFRSF1A, IL1R1, TNFAIP3, PLK1, PPP4C, TIFAB - returned exactly one reviewed row with a primary-name match.
UniProtKB REST search API, gene_exact queries with fields=accession,id,gene_primary,gene_synonym,protein_name,length, run 2026-09-20 · UniProt Consortium · UniProt 2026What this evidence cannot show
A single-row return with a primary match is strong but not proof of the right protein; it is why each protein entry in this folder asserts its accession and length for verify-accessions.py to check. Isoform-level ambiguity was not examined for any symbol.
Evidence that argues against this (1)
- A reviewed database record fixes the canonical sequence and primary symbol on the date it was read, and nothing more. It does not establish that the protein is expressed in KMT2A-rearranged blasts, which isoform is present, or that the subunit acts in this disease. Accession and length are asserted so that a swap fails verify-accessions.py - a defence this graph needs because IKK-family and NF-kB subunit names collide heavily in the older literature - not because the record carries biological weight. Xiu Y et al. 2018
- E3strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.database
Counted absences, all run 2026-09-20 against PubMed with querytranslation inspected. (KMT2A OR MLL OR "mixed-lineage leukemia") AND: IKBKG/NEMO/IKK-gamma returns 0; c-Rel spellings return 0; MAP3K7 OR TAK1 returns 3, all off topic; TRAF2 spellings return 2, both off topic; MYD88 returns 7, all off topic; TNFRSF1A OR TNFR1 returns 2, neither KMT2A-specific; NFKBIB OR NFKBIE OR IkappaB beta OR IkappaB epsilon returns 2, both chronic lymphocytic leukaemia. Separately, (KMT2A OR MLL) AND infant[tiab] AND NF-kB spellings returns 0, and (menin OR MEN1 OR revumenib OR ziftomenib) AND (IKK OR NF-kappaB inhibitor OR bortezomib OR IKK inhibitor) returns 3 with all three removed by title-checking.
PubMed E-utilities esearch and esummary, db=pubmed, run 2026-09-20 · NCBI · PubMed 2026What this evidence cannot show
These bound FRAMED literature, not measurement. The founding paper measures four of the enumerated genes and is reachable from none of their symbols, which is direct proof that a subunit-keyed count under-reports a complex's literature. Retired and case-variant aliases listed in the search-hygiene section were not run, and would raise several counts.
Evidence that argues against this (1)
- A counted absence keyed to a subunit symbol is systematically wrong for a complex. The founding paper of this field knocks down Ikka, Ikkb, Ikkg AND IRAK3 and is reachable from none of those four symbols, because its title and abstract frame at the level of 'IKK/NF-kB'. So every zero in this table bounds framed literature and understates measurement. Kuo HP et al. 2013
- E4established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Bounded rejection with a stated reversal condition: the only on-topic TRAF6 record in this disease is a paper in which IMPDH inhibition induces overactivation of Toll-like receptor-TRAF6-NF-kB signalling and upregulation of VCAM1, contributing to suppression of MLL-fusion leukaemia, with a TLR1/2 agonist used as an effective combination partner. TRAF6 is a named waypoint rather than a manipulated component, so no TRAF6 entry was written.
IMPDH inhibition activates TLR-VCAM1 pathway and suppresses the development of MLL-fusion leukemia. · Liu X, ..., Goyama S · EMBO Molecular Medicine 2023What this evidence cannot show
Abstract read; full text not retrieved. This paper is already cited in this graph by cell-state-innate-immune-lamp5, so the subtraction rule was applied before writing: it is not a new citation and the beat did not claim it as one.
Evidence that argues against this (1)
- Rejecting TRAF6 on the basis that its only on-topic record is an IMPDH paper is a statement about indexing, not about biology: in that paper the TLR-VCAM1 axis is the mechanism by which IMPDH inhibition suppresses MLL-fusion leukaemia, and TLR signalling runs through TRAF6. The node may be load-bearing precisely where the symbol search cannot see it. Liu X et al. 2023
this entry adjudicated the nine NF-kB targets with queries and counts, in prose and table rows
the enumeration's method finding: subunit-keyed absence counts under-report complex-level literature
zero records for IKBKG/NEMO in this disease, with the query
zero records for c-Rel in this disease, with the query
bounded rejections of MYD88, TAK1, TRAF2, TRAF6 and TNFRSF1A with their reversal conditions
no published combination of a menin inhibitor with any NF-kB-directed agent in any system