DDX3X couples RNA recognition and ATP hydrolysis to remodeling of short RNA duplexes. Human full-length protein and engineered catalytic fragments have been tested, but fragment activity does not establish the function of the naturally shorter product. The 2011 biochemical article has a 2026 correction that resolves several presentation errors while leaving specified original images and quantitative data unavailable; independently corroborated RNA chemistry remains central, and its substrate-dependent DNA result remains non-core with that limitation explicit. PMID:21589879(https://pubmed.ncbi.nlm.nih.gov/21589879/), PMID:29037760(https://pubmed.ncbi.nlm.nih.gov/29037760/), PMID:31300642(https://pubmed.ncbi.nlm.nih.gov/31300642/), PMID:41587182(https://pubmed.ncbi.nlm.nih.gov/41587182/).
An auxiliary translation-initiation role complements RNA helicase activity. DDX3X supports joining of mature subunits into functional 80S complexes, associates with initiation factors, and regulates selected structured transcripts. RNA dependence differs among the associations; effects on cyclin E1, ATF4 and growth are conditional and need not be uniform across cell types. PMID:22323517(https://pubmed.ncbi.nlm.nih.gov/22323517/), PMID:20837705(https://pubmed.ncbi.nlm.nih.gov/20837705/), PMID:29062139(https://pubmed.ncbi.nlm.nih.gov/29062139/).
Two signaling activities are mechanistically separable from RNA catalysis. DDX3X helps recruit IRF3 to IKKε during antiviral signaling, and it stimulates CK1 in Wnt signaling. The IRF3-recruitment mutant evidence and purified CK1 assays support these roles without assigning kinase activity to DDX3X, an experimentally resolved higher-order kinase architecture or a universal dependence on helicase activity. PMID:23478265(https://pubmed.ncbi.nlm.nih.gov/23478265/), PMID:29222110(https://pubmed.ncbi.nlm.nih.gov/29222110/), PMID:23413191(https://pubmed.ncbi.nlm.nih.gov/23413191/).
DDX3X shuttles between nucleus and cytoplasm, associates with translating particles and can enter stress granules. Cytosol/nucleoplasm observations refine broad locations; a peripheral 40S association does not make it a permanent ribosomal protein. Granule, exosome and mitochondrial observations describe distinct contexts rather than an exclusive compartment or extracellular catalytic function. PMID:15507209(https://pubmed.ncbi.nlm.nih.gov/15507209/), PMID:18596238(https://pubmed.ncbi.nlm.nih.gov/18596238/), PMID:28733330(https://pubmed.ncbi.nlm.nih.gov/28733330/), PMID:28869602(https://pubmed.ncbi.nlm.nih.gov/28869602/).
Mouse macrophage knockout/rescue and NLRP3-association experiments support a conserved contextual inflammasome role. The human ortholog inferences preserve species and construct limits. Poly(A)-tail contact, source-specific annealing, cadherin binding, two intrinsic-apoptosis assignments and the source-specific membrane/HCV-network observations remain unresolved where the necessary assays could not be independently adjudicated. Uncertainty preserves curator evidence and is not a negative biological assertion. PMID:31511697(https://pubmed.ncbi.nlm.nih.gov/31511697/), PMID:27546789(https://pubmed.ncbi.nlm.nih.gov/27546789/), PMID:25468996(https://pubmed.ncbi.nlm.nih.gov/25468996/), PMID:21883093(https://pubmed.ncbi.nlm.nih.gov/21883093/), PMID:16301996(https://pubmed.ncbi.nlm.nih.gov/16301996/).
The official products contain 662 and 646 residues. The shorter sequence replaces canonical residues 35–51 with a serine. No inspected experiment establishes identical activity or inactivity of that natural product; the 132–607, 168–582 and other laboratory fragments are separate constructs. Disease-variant reporter phenotypes likewise do not establish one universal gain-of-function mechanism. PMID:31300642(https://pubmed.ncbi.nlm.nih.gov/31300642/), PMID:20862261(https://pubmed.ncbi.nlm.nih.gov/20862261/), PMID:26235985(https://pubmed.ncbi.nlm.nih.gov/26235985/).
Positive generic binding assertions are retained as contextual when a specific partner class or mechanism cannot be justified. Supported class refinements do not imply that DDX3X activates the partner, performs its catalysis, or binds every site of a full-length product. This follows the current project annotation policy. No NEW row is proposed: the four core activities and their supported processes are already represented, and the two inspected GO-CAM models already encode the signaling-adaptor role.
Genuine Falcon authentication and configured fallback quota failures prevented a provider report; manual source research was performed instead. Original source records, publication caches and official product metadata remain unchanged. Access distinctions, literal provenance and assay limits are recorded in the evidence guide and primary-source notes.
The broad innate immune-response observation is retained as contextual: pattern-recognition signaling and regulation of interferon production do not logically entail GO:0045087 through current is_a/part_of ancestry. The source retains its positive TBK1/IRF3 response evidence. PMID:18583960(https://pubmed.ncbi.nlm.nih.gov/18583960/).
The auxiliary initiation-factor core is supported by the purified-component 80S experiment in PMID:22323517(https://pmc.ncbi.nlm.nih.gov/articles/PMC3367175/), Figure 7F–G, in addition to depletion and association experiments. GO:0003743 describes function in initiation; the demonstrated assembly contribution fits that definition. ATPase dependence and a molecular bridge remain hypotheses for this step. The complete bounded comparator query includes initiation-factor annotations for EIF5B and EIF4A1 but none for DHX29 or DHX33; that mixed convention is disclosed and does not justify adding an annotation merely to fill a gap.
The plasma-membrane assertion is retained contextually using independent human-cell imaging in PMID:29899501(https://pmc.ncbi.nlm.nih.gov/articles/PMC5998047/), with the original PMID:23413191 localization assay still unrecovered. The positive CAP-Rf reporter endpoint in PMID:10074132(https://pmc.ncbi.nlm.nih.gov/articles/PMC104042/) retains its unresolved causal stage; the NAS cyclin-E1 summary in PMID:22492871(https://pmc.ncbi.nlm.nih.gov/articles/PMC3387391/) is refined using established translation regulation. These judgments preserve the distinct source claims.
The current is_a/part_of graph confirms GO:0002753 → GO:0002221 → GO:0002758, but does not connect those terms to GO:0045087 by the proposed final edge. GO:0002758 directly inherits from GO:0002218 and GO:0002757. Regulation of an innate response is not substituted for the response itself; the experimentally supported TBK1/IRF3 context remains non-core PMID:18583960(https://pubmed.ncbi.nlm.nih.gov/18583960/).
CSB/ERCC6 and DHX33 associations are refined to ATPase binding on their established enzymatic class, preserving reciprocal co-immunoprecipitation and RNA-dependence limits PMID:26030138(https://pmc.ncbi.nlm.nih.gov/articles/PMC4451243/), PMID:26100019(https://pmc.ncbi.nlm.nih.gov/articles/PMC4525307/). This does not require DDX3X to activate the partner. Forty-five prior general terms now use MODIFY with already supported specific replacements, as the project instructions require. All 207 guide decisions and replacements were cross-checked against the final review; source identities, qualifiers, products and the 2011-paper correction assessment are preserved.
Nuclear-pore association and nuclear speckles are retained as distinct observations: the complete abstracts of PMID:15507209 and PMID:10329544 explicitly identify these locations. The latter also describes HCV-core-associated perinuclear redistribution. The compartment replacements preserve these observations rather than substituting nucleoplasm or cytosol alone; they do not assert exclusive residence or constitutive pore membership.
The diffuse cytoplasmic pool remains alongside condition-dependent stress granules in the evidence from PMID:18596238, PMID:21883093, PMID:29062139 (Results and Figure 1C) and PMID:24965446 (Results and Figure 3B). PMID:21883093 explicitly excludes processing bodies, and the DCP1a observations in PMID:24965446 are not transferred to DDX3. Endogenous DDX3X at the leading edge of spreading MRC5 fibroblasts is preserved separately from soluble cytosol (PMID:28733330, fractionation Results and Figure 4). Cell leading edge is not a typed descendant of cytoplasm in the inspected current ontology; the two replacements express complementary observed locations.
MAVS/IPS-1-associated mitochondrial proximity and HCV-associated lipid-droplet proximity remain qualified observations (PMID:20127681; PMID:21170385, Figures 6–7), not evidence of matrix import or of a particular membrane face. No new localization assertion is inferred from generic stress-granule background or from a partner's localization. These refinements change no action, imported source object, product, core function, or quotation.