Dyskerin performs two complementary kinds of work. Its conserved catalytic domain converts uridine to pseudouridine in RNA within H/ACA guide particles. Human DKC1 perturbation reduces rRNA modification and selected mRNA sites; the guide-dependent rRNA mechanism differs from the guide-independent term originally attached to one source. The conserved snRNA activity has direct yeast U2 guide-RNA, nuclease/rescue and catalytic-mutant evidence, with human H/ACA scaRNP membership providing a compatible target context. The yeast experiments are not human enzyme assays. PMID:25219674(https://pubmed.ncbi.nlm.nih.gov/25219674/), PMID:15962000(https://pubmed.ncbi.nlm.nih.gov/15962000/), PMID:19179534(https://pubmed.ncbi.nlm.nih.gov/19179534/).
In telomerase, dyskerin binds the H/ACA region of hTR and contributes its structure and stability. Human active-holoenzyme purification and cryo-EM distinguish the dyskerin-containing H/ACA units from the TERT catalytic lobe. Dyskerin depletion accelerates hTR loss, and inhibition of decay pathways can restore RNA and telomerase activity. Thus whole-complex activity and a contributes_to molecular-function claim are justified, while autonomous reverse-transcriptase activity is not. The older antisense paper remains abstract-only; the distinction rests on the independently established mechanism, not on silence in that abstract. PMID:29695869(https://pubmed.ncbi.nlm.nih.gov/29695869/), PMID:26950371(https://pubmed.ncbi.nlm.nih.gov/26950371/), PMID:12135483(https://pubmed.ncbi.nlm.nih.gov/12135483/).
The canonical official product is 514 residues. O60832-2 is named isoform 3 and is 420 residues, sharing residues 1–419 then ending in Arg420. The reported cytoplasmic variant is retained without assuming that preserved Asp125/PUA sequence proves catalytic equivalence or inactivity. Historical endogenous and tagged constructs are not automatically assigned to a current natural isoform. Nuclear dyskerin occupies nucleolar, nucleoplasmic and Cajal-body pools; the nucleolar snoRNP and Cajal-body scaRNP contexts belong to one RNA-isomerization mechanism, not to a claim that every substrate is modified in every listed compartment. PMID:21820037(https://pubmed.ncbi.nlm.nih.gov/21820037/), PMID:10556300(https://pubmed.ncbi.nlm.nih.gov/10556300/), PMID:23726835(https://pubmed.ncbi.nlm.nih.gov/23726835/).
Literal PAINT reconstruction places all six ancestral assertions on the human lineage. The target sequence matches the 514-residue official record within the 277-member tree/alignment. Experimental grounding includes human, yeast, rat and mouse contexts; a target appearing among descendant sources is valid grounding. The yeast snoRNA maturation phenotype is a structural assembly/quality-control role and is not a claim that dyskerin performs RNA cleavage. PMID:15964797(https://pubmed.ncbi.nlm.nih.gov/15964797/), PMID:9472021(https://pubmed.ncbi.nlm.nih.gov/9472021/).
Protein associations are judged by their actual evidence. RUVBL1 has recombinant dyskerin-binding support, permitting an ATPase partner-class refinement without assigning ATP hydrolysis to dyskerin. In contrast, the DHX36 association depends on TERC and the Npro capture is an RNP-proteome observation with an explicit indirect-association possibility. Both remain positive non-core associations. NAF1/SHQ1 assembly functions and TCAB1 RNA-chaperone activity do not automatically establish the protein-folding-chaperone definition. PMID:18358808(https://pubmed.ncbi.nlm.nih.gov/18358808/), PMID:21846770(https://pubmed.ncbi.nlm.nih.gov/21846770/), PMID:24965446(https://pubmed.ncbi.nlm.nih.gov/24965446/), PMID:19383767(https://pubmed.ncbi.nlm.nih.gov/19383767/).
The human UV-capture study supports mRNA binding through its exact DKC1 Table S1 row and stringent RNA-protein crosslinking protocol. This does not identify the poly(A) tail as the contact site. The imported long RNAcentral telomerase-RNA record contains the canonical 451-nt sequence, but no historical construct endpoint is inferred from that database record. PMID:22658674(https://pubmed.ncbi.nlm.nih.gov/22658674/), PMID:18082603(https://pubmed.ncbi.nlm.nih.gov/18082603/), PMID:22527283(https://pubmed.ncbi.nlm.nih.gov/22527283/).
Two source-specific localization claims remain unresolved: DKC1-dependent scaRNA localization and TCAB1-protein localization from the 2014 TRiC paper. The accessible main text directly supports TERC abundance/localization and telomerase recruitment, but those results cannot be substituted for the distinct supplemental endpoints. Failed supplemental downloads were HTML challenges or access errors, not inspected experiments. These positive curator assertions are preserved as UNDECIDED rather than declared wrong. PMID:25467444(https://pubmed.ncbi.nlm.nih.gov/25467444/).
The normal Falcon request failed with an authorization error and its configured fallback failed with a quota/authentication error. No provider report was fabricated. The source guide and these notes document manual evidence review. Generic positive interactions are retained under the scoped project instruction. Full-text flags reflect the unmodified normal cache; separately recovered original Methods/Results and exact access limits are identified in the source guide.
The broad RNA-processing refinements preserve both rRNA processing and the inherited snRNA pseudouridylation role, grounded by the actual Cbf5 guide-dependent U2 experiments. Current typed ancestry does not place mRNA pseudouridylation beneath RNA processing, so that established role remains separate. Cajal body is part of nucleoplasm; the nuclear replacement preserves diffuse and assembled pools without inserting a redundant nested location. PMID:15962000(https://pubmed.ncbi.nlm.nih.gov/15962000/).
The broad rRNA-pseudouridylation annotations retain their valid PAINT ancestry and mammalian rule support, but GO:0000454 specifies the snoRNA-guided target mechanism already established by human DKC1/H/ACA evidence and mammalian guide-dependent particles (PMID:25219674(https://pubmed.ncbi.nlm.nih.gov/25219674/), PMID:12446766(https://pubmed.ncbi.nlm.nih.gov/12446766/)). This refinement does not rewrite the broader ancestral IBD or treat self-grounding as circular.
The mouse experiment in PMID:15240872(https://pubmed.ncbi.nlm.nih.gov/15240872/) concerns A353V and G402E dyskerin in embryonic stem cells, not RNP reconstitution. Both mutations alter pseudouridylation, subsets of H/ACA snoRNAs and pre-rRNA processing; the strong telomerase-RNA and telomerase-activity defect applies to A353V, not G402E. The recombinant yeast RNP reconstitution and ternary structure instead belong to PMID:22085967(https://pubmed.ncbi.nlm.nih.gov/22085967/). These corrected findings are supported by the complete cached abstracts; bounded original-recovery attempts did not recover full text.
The nine Reactome references now distinguish assembly, recruitment, template alignment, extension, translocation, dissociation, guide-directed pre-rRNA modification and SHQ1 binding. Their positive paths retain DKC1 residues 2–514 in nucleoplasm, without assigning TERT polymerase chemistry or an autonomous translocation/dissociation activity to dyskerin. See the individual public reaction links in the source-evidence guide.
The inspected is_a/part_of closure still places mRNA pseudouridine synthesis under mRNA modification and pseudouridine synthesis, both under RNA modification; it does not supply an edge to RNA processing. The existing rRNA-processing and snRNA-pseudouridylation refinements therefore remain unchanged. The telomerase core retains nucleoplasm, which includes the separately supported Cajal-body pool, without asserting exclusive localization or adding nested locations solely to duplicate accepted annotations.