GeneweakQ136-interactome-tail-size-and-abundance

DDX55 - a large-subunit rRNA helicase that is nucleolar by imaging and nucleoplasmic by its own founding paper, and the narrowest RBP bait to reach KMT2A after DDX52

also called DEAD-box helicase 55 · KIAA1595 · DExD box ATPase DDX55

DDX55 is an enzyme that unwinds RNA while the big half of the ribosome is being built. Two different kinds of measurement disagree about where in the nucleus it sits - antibody imaging says the nucleolus, its own discovery paper says the nucleoplasm - and that disagreement matters, because whether this project's nucleolar interactome signal is real depends on which proteins count as nucleolar.

The same thing, for a scientist

DDX55 (HGNC:20085, ENSG00000111364, UniProt **Q8NHQ9**, 600 aa, 12q24.31, alias KIAA1595) is a DExD-box ATPase of large-subunit ribosome biogenesis: it associates with pre-ribosomal subunits, is required for maturation of LSU pre-rRNAs, selectively binds double-stranded RNA, and crosslinks to **helix H62 in domain IV of the 28S rRNA**, with a nuclear localisation signal and pre-ribosome recruitment determinant both in its C-terminal region [E4]. **Its founding paper's headline is a localisation claim**: *"in contrast to many other ribosome biogenesis factors, the DExD box ATPase DDX55 predominantly localizes to the nucleoplasm"* [E4]. **The Human Protein Atlas disagrees**, giving DDX55 a main location of **nucleoli** with nucleoplasm and cytosol additional, at the *Enhanced* reliability tier [E5]. It enters this project on one row - IntAct **IM-30059**, **MI:1314 BioID2**, **MI:1060 spoke**, **DDX55 bait / KMT2A prey**, MI-score 0.27, K562 [E1][E2] - and DDX55 is one of the **narrowest baits in that panel**: 131 preys against a panel median of 326, ranking 45th of 50, giving **p = 0.162** and a verdict of NOT RESOLVED [E3]. **The reason to write it is that its nucleolar status was an explicit boundary case in this graph's own nucleolar-enrichment test**, and the HPA call settles which side of the boundary it falls on.

What it looks like

DDX55 - a large-subunit rRNA helicase that is nucleolar by imaging and nucleoplasmic by its own founding paper, and the narrowest RBP bait to reach KMT2A after DDX52600 aa

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.

Go deeper

The subnuclear disagreement, which is the substance of this entry

ana-q129-the-nucleolar-triple-survives-none-of-the-four-pre-registered-discriminators asked whether the nucleolar proteins in the K562 RBP panel are enriched among the baits reaching KMT2A and PSIP1. Its classification called 9 of 50 baits nucleolar or ribosome-biogenesis (DDX21, DDX52, GNL3, NOLC1, NPM1, RPS3, ...), and stated explicitly: "a generous version adding DDX55 and EXOSC5 gives 11 of 50 = 22%, against which the observed 22% is an exact match, so the conclusion is insensitive to the boundary." DDX55 was named as the boundary case and the boundary was never resolved.

This entry resolves it, in both directions, because the two resources genuinely conflict:

source	call	tier
Choudhury et al. 2021, the DDX55 paper	"predominantly localizes to the nucleoplasm", in contrast to other ribosome biogenesis factors	primary experiment, the paper's own headline
UniProt Q8NHQ9	Nucleus, nucleoplasm, ECO:0000269 coded to that same paper	curated, coded, consistent with the paper
HPA v25.1 immunofluorescence	Nucleoli main; nucleoplasm and cytosol additional	Enhanced reliability, the top imaging tier
HPA function annotation	ribosome biogenesis, rRNA processing, rRNA-binding	gene-level

Both are right about different things. DDX55 is functionally a ribosome-biogenesis factor that binds 28S rRNA, which is nucleolar work, and its steady-state pool is reported nucleoplasmic by the group that characterised it. For the purposes of the cycle-129 test the generous classification is the defensible one - the resource with the top imaging tier calls it nucleolar and the function annotation agrees - and that analysis already reported that its conclusion is unchanged either way. So this entry confirms a prior analysis rather than overturning it, and the confirmation is the point: the boundary case had a resolvable answer that nobody had fetched.

The interaction row

One IntAct row, IMEx IM-30059: MI:1314 BioID2, MI:1060 spoke, roles MI:0496 bait (DDX55) / MI:0498 prey (KMT2A), host taxid:9606(human-k_562), MI-score 0.27, the deposition-wide floor 2.

DDX55 recovered 131 of 2,681 preys - the panel median is 326 and DDX55 ranks 45th of 50, making it one of the four narrowest baits in the map. That selectivity is why its p is as low as it is: holding both the bait's reach and KMT2A's recoverability (25 of 50 baits), p = 0.162, NOT RESOLVED 3. Only DDX52 (166 preys, p = 0.201) and ZC3H8 (152, p = 0.185) are comparable among the RBP baits, and all three fall on the wrong side of the pre-registered PROMOTED threshold.

It is one spoke-expanded proximity row with no orthogonal validation, in a cell with no KMT2A rearrangement, and the entry does not inflate it. What it adds over the row is the localisation adjudication above and the fact that KMT2A has a documented nucleolar pool in this graph - absence-kdm2b-and-the-kmt2a-fusion-are-both-nucleolar-and-nobody-has-put-them-together - so a nucleolar bait reaching KMT2A is at least compartment-compatible, which is a statement about plausibility and not about evidence.

Localization, cell type, epigenetics, and a bounded absence

Subcellular (UniProt): Nucleus, nucleoplasm, ECO:0000269 coded to PMID 33048000 - and here the pointer and the fact coincide, because that paper's abstract leads with the localisation result 4.

Subnuclear (HPA): nucleoli main; nucleoplasm and cytosol additional; reliability Enhanced 5. See the section above.

Domain architecture: helicase ATP-binding domain 40-223, helicase C-terminal domain 254-402, a disordered stretch 500-550, and a region annotated "Important for nuclear localization" at 533-562 4. The C-terminal region carries both the NLS and the determinant for pre-ribosome recruitment and dsRNA selectivity.

Cell type of the observation: K562 chronic myeloid leukaemia (BCR::ABL1, no KMT2A rearrangement). Cell state: over-expressed BioID2 fusion construct, one isoform per protein, proliferating line. Cell type of the biology: HEK293 and HeLa, in the characterisation paper.

Epigenetic role: none. DDX55 is an RNA helicase acting on pre-rRNA; it has no chromatin-modifying, chromatin-reading or DNA-binding activity.

Other fusion families, bounded absence, both operands 6:

query	translation	n
`DDX55[tiab] AND KMT2A[tiab]`	`"DDX55"[Title/Abstract] AND "KMT2A"[Title/Abstract]`	0
`DDX55[tiab] AND MLL[tiab]`	`"DDX55"[Title/Abstract] AND "MLL"[Title/Abstract]`	0
`DDX55[tiab]` (whole literature)	-	13

Both zeros are genuine - non-empty translation, warninglist reading No items found. The whole DDX55 literature is 13 papers, so the absence says the field has not looked rather than that the relationship was sought and not found. HGNC lists no previous symbol; the alias KIAA1595 was not queried. Other fusion families: nobody has looked at DDX55 in any of them.

Facts and figures

hgnc ID	HGNC:20085
symbol	DDX55
ensembl gene ID	ENSG00000111364
uniprot	Q8NHQ9
length AA	600
chromosomal location	12q24.31
entrez ID	57696
identifier provenance	HGNC REST fetch/symbol/DDX55 and UniProtKB Q8NHQ9 retrieved live 2026-09-23; Ensembl id taken from the HGNC record. HGNC lists no previous symbol. DDX55 is named in 5 node files by word-boundary match - the KMT2A promiscuity census, the nucleolar-triple analysis, the read-queue enumeration and two others - and no entry existed.
subnuclear location hpa	Nucleoli (MAIN), with nucleoplasm and cytosol additional, reliability Enhanced
subnuclear conflict	THE TWO SOURCES DISAGREE. The characterisation paper's headline is 'predominantly localizes to the nucleoplasm, in contrast to many other ribosome biogenesis factors'; HPA's top-tier imaging call is nucleoli. Both are recorded. For the cycle-129 nucleolar classification the generous reading (DDX55 = nucleolar) is the defensible one, and that analysis already stated its conclusion is insensitive to the boundary.
cell type of the observation	K562 chronic myeloid leukaemia - a leukaemia cell, but BCR::ABL1 and not KMT2A-rearranged
cell state of the observation	over-expressed BioID2 fusion construct, one isoform per protein, proliferating line
cell type of the biology	HEK293 and HeLa, in the characterisation paper; no haematopoietic system
epigenetic role	None. DDX55 is an RNA helicase acting on pre-rRNA with no chromatin-modifying, chromatin-reading or DNA-binding activity.
shown for fusion	NO - nobody has looked in any fusion. The row is wild-type KMT2A in K562, and PubMed returns zero for DDX55 against both KMT2A and MLL.
evidence class	single publication, single method, spoke-expanded BioID2 proximity, no orthogonal validation - written to resolve a named localisation boundary case, not for the row

domains

prev and alias symbols

[
  "KIAA1595"
]

domains — the numbers behind the chart

[
  {
    "name": "Helicase ATP-binding",
    "start": 40,
    "end": 223
  },
  {
    "name": "Helicase C-terminal",
    "start": 254,
    "end": 402
  },
  {
    "name": "Region important for nuclear localization",
    "start": 533,
    "end": 562
  }
]

interaction with project baits

[
  {
    "core_protein": "KMT2A",
    "core_uniprot": "Q03164",
    "partner_experimental_role": "bait",
    "core_experimental_role": "prey",
    "method": "MI:1314 proximity-dependent biotin identification (BioID2)",
    "expansion": "MI:1060 spoke",
    "mi_score": 0.27,
    "imex": "IM-30059",
    "pmid": "39251607",
    "host_cell": "K562 (BCR::ABL1 CML; no KMT2A rearrangement)",
    "n_rows": 1,
    "n_publications": 1
  }
]

adjudication

{
  "verdict": "NOT RESOLVED",
  "p_pair_both_margins_held": 0.162,
  "bait_preys_recovered": 131,
  "panel_median_preys_per_bait": 326,
  "bait_selectivity_rank_in_panel": "45 of 50 (1 = widest labeller)",
  "prey_baits_recovering_it": 25,
  "of_panel": 50,
  "comparable_narrow_baits": {
    "ZC3H8": 0.185,
    "DDX52": 0.201
  },
  "why_written_anyway": "DDX55 was the named BOUNDARY CASE in ana-q129-the-nucleolar-triple-survives-none-of-the-four-pre-registered-discriminators ('a generous version adding DDX55 and EXOSC5 gives 11 of 50'), and the boundary had a resolvable answer nobody had fetched",
  "source": "analyses/interactome-tail-size-and-abundance-adjudication/pair_null.json"
}

subcellular location uniprot

[
  "Nucleus, nucleoplasm (ECO:0000269, cited to PMID 33048000 - the pointer and the fact coincide)"
]

bounded absence

{
  "queries": [
    "DDX55[tiab] AND KMT2A[tiab] -> 0",
    "DDX55[tiab] AND MLL[tiab] -> 0"
  ],
  "querytranslations": [
    "\"DDX55\"[Title/Abstract] AND \"KMT2A\"[Title/Abstract]",
    "\"DDX55\"[Title/Abstract] AND \"MLL\"[Title/Abstract]"
  ],
  "name_list_tried": [
    "DDX55"
  ],
  "name_list_NOT_tried": [
    "KIAA1595"
  ],
  "total_literature": "13 records for DDX55 in any title or abstract",
  "bound": "titles and abstracts only; supporting figures not searched"
}

How this could bite us back

DDX55 is annotated by HPA as an essential protein and its job is large-subunit ribosome biogenesis, which every proliferating cell needs. Any intervention against it is a ribosome-biogenesis intervention, and this project already tracks the nucleolar-block toxicity axis for exactly that reason.
If the KMT2A-DDX55 proximity is a shared nucleolar address rather than an association, then reading it as evidence of anything would push the graph toward a compartment-driven interactome that is mostly a map of where proteins live. That failure mode is the single most likely way this beat's promotions turn out to be wrong.

Still open

Which localisation is right for DDX55 in a haematopoietic cell? Neither the characterisation paper (HEK293, HeLa) nor HPA measured it in blood, and the nucleolar-versus-nucleoplasmic question is exactly the kind that changes with cell type and growth rate. A single immunofluorescence in a KMT2A-rearranged line would settle it and would also settle the classification boundary in this graph's own nucleolar-enrichment test.
Does KMT2A's documented nucleolar pool overlap DDX55's? This graph carries an entry noting that KDM2B and the KMT2A fusion are both nucleolar and nobody has put them together. DDX55 is a third nucleolar protein now linked to KMT2A by a proximity row, and the shared-compartment reading is the deflationary one - but it has never been tested by co-staining.
Is 28S rRNA helix H62 remodelling relevant to anything in leukaemia? DDX55's one mechanistic result is a specific crosslink site in LSU biogenesis, and ribosome biogenesis is a live axis in this graph through the nucleolar block entries. Whether large-subunit maturation is differentially required in KMT2A-rearranged cells is unmeasured.
The three narrowest RBP baits reaching KMT2A - DDX55, ZC3H8, DDX52 - are the three with the lowest p, and two of the three are ribosome-biogenesis proteins. Is that a compartment signal or an artefact of narrow baits being narrow because they label a small compartment? A bait-width-matched comparison against non-nucleolar narrow baits would separate them.

Every source behind this entry (6)

E1established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.paper
Khoroshkin et al. fused BioID2 to 50 human RNA-binding proteins in K562 cells and deposited the curated pair list as IMEx IM-30059; DDX55 is one of the 50 baits and KMT2A is among its recovered preys.
Systematic identification of post-transcriptional regulatory modules. · Khoroshkin M, Buyan A, Dodel M, Navickas A, Yu J, Trejo F, et al. · Nat Commun 2024
PMID 39251607 DOI
What this evidence cannot show
Abstract read; full text not read. The authors state that RBP over-expression, BioID2 fusion and the use of a single isoform per protein may bias the data. K562 carries BCR::ABL1 and has no KMT2A rearrangement.
Evidence that argues against this (1)
- KMT2A is recovered by 25 of these 50 baits, and this graph has traced that breadth largely to KMT2A's size. Being one of 25 is close to a coin flip and carries about one bit. source 2026
E2established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
IntAct serves the DDX55-KMT2A record as PSI-MI MI:1314 proximity-dependent biotin identification, MI:1060 spoke expansion, DDX55 experimental role MI:0496 bait and KMT2A MI:0498 prey, MI-score 0.27, host taxid:9606(human-k_562), IMEx IM-30059, publication PMID 39251607.
IntAct molecular interaction database (EMBL-EBI), PSICQUIC MITAB 2.7, pubid:39251607, retrieved 2026-09-23 2026
What this evidence cannot show
Retrieved live 2026-09-23; 19,403 rows fetched against an endpoint-declared count of 19,403. Rule 93: PSICQUIC is a floor relative to IntAct REST and a REST query on this pair was not run. Spoke expansion means the binary pair was never measured.
Evidence that argues against this (1)
- MI-score 0.27 is the deposition-wide value on every row in this file and carries no pair-specific information. source 2026
E3moderate confidence 3 of 5: One strong study, or several smaller consistent ones.dataset
In IM-30059 DDX55 recovered 131 of 2,681 human preys, against a panel median of 326, ranking 45th of 50 baits by reach; under a null holding both the bait's reach and the prey's recoverability, DDX55 recovering KMT2A has p = 0.162, which is NOT RESOLVED under the pre-registered thresholds, with ZC3H8 (152 preys) at 0.185 and DDX52 (166) at 0.201.
analyses/interactome-tail-size-and-abundance-adjudication/ - orientation.py, pair_null.py, adjudication.json (this repository), pre-registered in PREREGISTRATION.md 2026
What this evidence cannot show
20,000 draws, seed 20260923. A Bonferroni threshold over the 62 pairs tested is 0.0008 and nothing survives it; this is an ordering, not a test. The null assumes preys are drawn independently given their degrees, which is false when baits share a complex - and DDX52, DDX55 and ZC3H8 are all RNA-processing baits whose prey sets are likely correlated.
Evidence that argues against this (1)
- The three lowest-p RBP baits reaching KMT2A are the three narrowest. That is what the null is built to do, and it means the ordering is largely an ordering of labelling breadth rather than of association. source 2026
E4strong confidence 4 of 5: Replicated in peer-reviewed work by more than one group.paper
In contrast to many other ribosome biogenesis factors, the DExD box ATPase DDX55 predominantly localises to the nucleoplasm, with a nuclear localisation signal in its C-terminal region; DDX55 associates with pre-ribosomal subunits in human cells and is required for maturation of large subunit pre-rRNAs, selectively associates with double-stranded RNA substrates which stimulate its ATPase activity, and crosslinks to a binding site in helix H62 of the 28S ribosomal RNA, identifying domain IV as a potential remodelling target. UniProt Q8NHQ9 is 600 aa with a helicase ATP-binding domain at 40-223, a helicase C-terminal domain at 254-402 and a region important for nuclear localisation at 533-562.
The DExD box ATPase DDX55 is recruited to domain IV of the 28S ribosomal RNA by its C-terminal region. · Choudhury P, Kretschmer J, Hackert P, Bohnsack KE, Bohnsack MT · RNA Biol 2021
PMID 33048000 DOI
What this evidence cannot show
Abstract read; full text not read. Work is in HEK293 and HeLa cells, neither haematopoietic. The UniProt domain boundaries were fetched live from rest.uniprot.org/uniprotkb/Q8NHQ9.json on 2026-09-23. The nucleoplasmic call is this group's own and is contradicted at the Enhanced tier by HPA imaging - see E5.
Evidence that argues against this (1)
- The Human Protein Atlas assigns DDX55 a MAIN location of nucleoli at its top reliability tier, with nucleoplasm only as an additional location. A protein that binds 28S rRNA and matures LSU pre-rRNA is doing nucleolar work whatever its steady-state pool looks like, so the 'predominantly nucleoplasmic' headline is a statement about a snapshot and not about where the function happens. source 2026
E5moderate confidence 3 of 5: One strong study, or several smaller consistent ones.database
The Human Protein Atlas assigns DDX55 a main subcellular location of nucleoli with nucleoplasm and cytosol as additional locations at the Enhanced reliability tier, classes it as an essential protein and an enzyme with molecular functions helicase, hydrolase, RNA-binding and rRNA-binding, and assigns it the biological processes ribosome biogenesis and rRNA processing.
Human Protein Atlas v25.1, proteinatlas.json.gz record for DDX55 (ENSG00000111364), retrieved 2026-09-23 2026
What this evidence cannot show
Antibody-based immunofluorescence in a small cell-line panel, not in leukaemic blasts. 'Enhanced' is HPA's top reliability tier but it is still one antibody's staining pattern in fixed cells. The call CONTRADICTS the DDX55 characterisation paper's own headline; this entry reports both rather than choosing silently.
Evidence that argues against this (1)
- The group that characterised DDX55 states the opposite as their lead result and frames it as a contrast with other ribosome biogenesis factors, which means they were looking specifically at this question. A database imaging call should not silently override a targeted primary result. source 2021
E6established confidence 5 of 5: Multiple independent lines of evidence; treated as settled in the field.database
PubMed returns zero records for DDX55 AND KMT2A in titles and abstracts and zero for DDX55 AND MLL, against a whole DDX55 literature of 13 records.
NCBI PubMed esearch via scripts/pubmed_safe.py, run 2026-09-23; querytranslations recorded: "DDX55"[Title/Abstract] AND "KMT2A"[Title/Abstract] and "DDX55"[Title/Abstract] AND "MLL"[Title/Abstract] 2026
What this evidence cannot show
Both zeros carry a non-empty querytranslation and a warninglist reading 'No items found.'. NAME LIST TRIED: DDX55 only. HGNC lists no previous symbol; the alias KIAA1595 was NOT queried and would be the spelling in pre-2005 sequencing literature. Bounded by titles and abstracts only.
Evidence that argues against this (1)
- A thirteen-paper literature makes this absence uninformative about biology. It records that nobody has looked, not that the relationship was sought and not found. source 2026

Connected to

related to →KMT2A / MLL1 — the gene that breaks
DDX55 bait / KMT2A prey, one MI:1314 BioID2 spoke row, IM-30059, K562; p = 0.162, NOT RESOLVED
supports →The ABT1/KRR1/UTP3 nucleolar triple: pre-registered, then adjudicated - it is not three proteins on two baits but a seven-member ribosomal block in one screen plus one hit in a neighbourhood that shows no nucleolar enrichment, and two of the four pre-registered discriminators were void exactly as the pre-registration predicted
resolves the named boundary case: HPA calls DDX55 nucleoli-main at Enhanced reliability, so the generous 11-of-50 classification is the defensible one
part of →The 50-RBP BioID2 proximity map in K562 - the first modern proximity dataset this graph can reach, and AFF1 is in it as prey
one of the 50 BioID2 baits in IM-30059, and the 45th of 50 by reach
related to →Three CXXC-pathway activities converge on rDNA - KDM2B represses Pol I, KMT2A and SETD1A bind the whole locus, MLL-AF4 is predominantly nucleolar - and no paper contains two of them
a third nucleolar protein now linked to KMT2A by a proximity row; the shared-address reading is the deflationary one and has never been tested by co-staining
supports →Prey recovery in the two BioID depositions tracks SIZE and not abundance - and the interactome tail was being scored on the wrong side of the pair, because every one of the 22 partners is the BAIT and every core protein is the PREY
one of the three narrowest RBP baits reaching KMT2A, all three NOT RESOLVED
co-occurs with →ZC3H8 / Fliz1 - the last little-elongation-complex subunit without an entry, discovered in fetal liver haematopoietic progenitors, and a selective BioID2 bait that reaches KMT2A in K562
two of the three narrowest baits in IM-30059 that reach KMT2A (131 and 152 preys, p = 0.162 and 0.185)
← co-occurs withPSIP1-FASTKD2 declined: the bait is a mitochondrial matrix and nucleoid protein with two experimental localisations, the prey is nucleoplasmic LEDGF, and the row is one spoke-expanded BioID2 proximity in K562
the two ribosome-biogenesis baits in IM-30059 that reach a core protein - one mitochondrial, one nucleolar, both NOT RESOLVED by the statistic