The review retains the 32 imported annotation records and six UniProt product identities. A newer QuickGO release has 34 rows: it changes the electronic mitochondrial reference and adds two electronic cytoplasm/cytosol assertions. Those are documented as a source-version difference; the older assertions have not been silently replaced. The review applies the project curation instructions: use supported informative replacements and retain positive generic interactions when specificity cannot be justified.
PMID:8692979(https://pmc.ncbi.nlm.nih.gov/articles/PMC38970/) was read from all five original scanned pages. Page 7259, Enzyme Assays and Table 1, reports the expressed human maltose-binding fusion after cleavage. dCyd phosphorylation is low but nonzero, about one tenth of dGuo under both tested substrate concentrations. The competition result supports low affinity. The 780-bp historical open reading frame must not be equated to every modern alternative product.
PMID:17073823(https://pmc.ncbi.nlm.nih.gov/articles/PMC1798436/), Subjects and Methods—Enzymatic measurements, and Results—DGUOK enzymatic measurements, describes dA and dG radiochemical assays in mitochondria-enriched human fibroblast preparations. Excess dC suppresses cytosolic DCK. PMID:23043144(https://pmc.ncbi.nlm.nih.gov/articles/PMC3501975/), Enzymatic measurements and Table 2, independently assays dG phosphorylation in patient/control muscle. These natural tissue preparations differ from recombinant proteins and do not establish activity of each short natural product.
The original PMID:8706825 abstract establishes recombinant human enzyme purification and purine-substrate/matrix context; its full methods were not recovered. For PMID:11427893(https://pubmed.ncbi.nlm.nih.gov/11427893/), the original abstract and published erratum, pages 818–819 were read. The erratum is not a substitute for the inaccessible original structural Methods and does not overturn the measured low dCyd activity.
The Reactome reaction R-HSA-74207 links catalyst activity 74206, DGUOK dimer R-HSA-74205 and Q16854 component R-HSA-53573. The component is explicitly in mitochondrial matrix. Its modeled residues 40–277 reflect the curated mature-chain model; they do not independently demonstrate cleavage or the behavior of short products. Broad mitochondrial/cytoplasmic source rows are refined with this independent matrix evidence, while retaining the source-specific assay limitations. Cytoplasm includes mitochondria; it is not a synonym for cytosol.
For PMID:21630459(https://doi.org/10.1002/pmic.201000799), the original abstract describes microscopy checks on purified sperm nuclei. The coauthor's deposited thesis, Appendix 8.2, explicitly reproduces the original protein catalogue. Table 8.1, printed page 198 (PDF page 199), lists DGUOK/Q16854 with 13.4% coverage and two peptides. This corrects the earlier contamination assumption. It establishes a specialized detection, not a demonstrated nuclear catalytic role. A later table in the thesis is a different dataset.
PMID:19221117(https://pubmed.ncbi.nlm.nih.gov/19221117/) and PMID:11687801(https://pubmed.ncbi.nlm.nih.gov/11687801/) remain abstract-only in this assessment. Their imported experimental annotations are not rejected merely because the accessible abstracts lack the original assay details. Independent human DGUOK chemistry and localization support the biological conclusions.
PMID:33961781(https://pmc.ncbi.nlm.nih.gov/articles/PMC8165030/), Clone Construction, Cell Culture and Affinity-Purification Mass Spectrometry, uses sequence-verified C-terminal FLAG-HA baits. Public IntAct records IM-29278-15667, IM-29278-63, IM-29278-28719 and IM-29278-27299 contain DCK/DGUOK associations in HEK293T/HCT116 cells; both participant sequence/taxon identifiers agree with canonical human proteins. The reciprocal HEK293T bait observation strengthens the association but is not purified binary binding. The 2025 U2OS study, AP–MS data collection and Data availability, releases the AP-MS network: edge 578496 links DCK node 561794 (bait) to DGUOK node 578494 (prey). The public network data preserve those identifiers. These are co-complex associations after ectopic bait expression; native mitochondrial heterodimerization and kinase activation are not demonstrated. GO:0019900 kinase binding fits the catalytic partner class; protein kinase binding would incorrectly imply that DCK phosphorylates protein substrates.
The actual PANTHER PTHR10513 tree places Q16854 beneath PTN002236991 (three substrate-activity IBDs), PTN002236975 (mitochondrion) and PTN000053758 (cytoplasm). Target experimental annotations used in the ancestral evidence are legitimate, not circular. Low dCyd activity is compatible with a secondary inherited activity.
The rat Dguok donor D3ZDE4 carries IMP evidence from PMID:18007665(https://doi.org/10.1038/sj.cdd.4402258) for the neurite process and PMID:30404003(https://pmc.ncbi.nlm.nih.gov/articles/PMC6289059/) for the respiratory phenotype. For the first paper, the abstract and public supplementary figure legends were read; the precise DGUOK rat-validation panel remains inaccessible. For the second, Figure 3 and Figure 7 plus Methods establish human knockout/rescue and rat respiratory effects. These data show that DGUOK supports mtDNA-dependent respiration through nucleotide supply; they do not make the kinase an electron-transfer component. The mouse source Q9QX60 and PMID:10455141(https://pubmed.ncbi.nlm.nih.gov/10455141/) distinguish mitochondrial full-length and cytoplasmic N-terminally truncated mouse proteins, without assigning those properties to human isoforms.
The six official human products reconstruct to 277, 189, 239, 151, 255 and 89 residues. Their identities and sequence-change fields are retained. Predicted targeting, disease truncations, historical recombinant constructs and natural splice products are distinct types of evidence. The two core entries describe measured substrate activities of one enzyme, with no equal-function assertion for shorter products and no additional annotation proposed.
Publication cache flags and original machine files are preserved. Separately recovered original pages, author-deposited material and released study datasets are distinguished above from abstract-only access. Automated research failed authentication/quota checks; no provider report was invented.
GO:0005524 is retained because ATP binding supports the phosphotransfer reaction. The kinase activities GO:0004136 and GO:0004138 are not descendants of this binding term. PMID:8692979(https://pmc.ncbi.nlm.nih.gov/articles/PMC38970/), printed page 7259, Enzyme Assays, specifies labeled ATP with the cleaved human recombinant protein. The original Results paragraph adjacent to Table 1 states: “Low levels of phosphorylated dCyd were also detected by this assay.” Table 1 reports relative dCyd phosphorylation of 0.1 ± 0.03 at 5 µM and 0.1 ± 0.05 at 100 µM, with dGuo normalized to 1.0. These original-page observations are separately accessible evidence; the publication cache remains abstract-only and has not been rewritten.
GO:0010977 describes negative regulation of neurite development, whereas GO:0042775 describes electron transfer. The accessible PMID:18007665 abstract explicitly names DGUOK among inhibitory screen hits; its supplementary figure legends, Figures S1/S2, describe human SH-SY5Y siRNA treatment and the neurite-length readout. This supports a contextual regulatory phenotype, without a defined molecular mechanism. The exact DGUOK rat-validation panel remains unrecovered; retention of the literal rat transfer includes curator deference. It does not imply that every perturbation phenotype proves process participation. By contrast, PMID:30404003(https://pmc.ncbi.nlm.nih.gov/articles/PMC6289059/), Results on DGUOK-deficient hepatocytes and Figure 3B–I, links nucleotide supply, mtDNA abundance and respiratory integrity. DGUOK does the nucleotide phosphorylation; respiratory complexes perform electron transfer.
The public Reactome model can be checked directly through reaction 74207, catalyst 74206, dimer 74205 and protein 53573. The literal chain is catalystActivity.dbId 74206 → physicalEntity.dbId 74205 → hasComponent.dbId 53573. The final object has referenceEntity.identifier Q16854, startCoordinate 40, endCoordinate 277, and compartment accession 0005759. This is a curated mature-chain model, consistent with the predicted targeting sequence; it is not a new cleavage assay.