The imported products remain Q5T890-4/name 3 (displayed, 1,561 residues), Q5T890-1/name 1 (1,550 residues) and Q5T890-2/name 2 (712 residues). Historical fragments and engineered constructs are distinguished from these natural products. Source records and normal publication caches are unchanged. The assessments below describe the original material actually inspected; availability metadata alone does not establish access to an experiment.
| Source | Inspected evidence and limits |
|---|---|
| PMID:37014751(https://doi.org/10.1016/j.celrep.2023.112329), original | Results Figures 1A–C (centromeres), 4D/F/J–N (PCNA binding), 5A–G (damage-end processing), with constructs, immunofluorescence, laser, pull-down, ITC and crystallography Methods. Gene-synthesized human 1,561-residue YFP protein is separate from C-terminal fragments and aPIP peptides. GST 753–819 and biotinylated 790–811 bind PCNA with Q798A controls; peptide 795–816 was used for ITC/crystallography. The authors limit functional PCNA-mutant interpretation beyond localization. The annotation reviewer read text/captions; an independent reviewer inspected Figures 1 and 4. Detailed supplementary experiments were not read. |
| PMID:32355287(https://doi.org/10.1038/s41422-020-0328-3), original | Results Figures 1–5 and cloning, laser and co-IP Methods; selected original Supplementary Table S2 cell/construct entries. Human U2OS recruitment/damage sensitivity and HEK293T association are separate from mouse B-cell class-switch recombination, XLF-redundant V(D)J joining and MRI/ligase recruitment. Benzonase/ethidium-bromide co-IP supports association without proving every pair binary. Table S2 distinguishes human constructs ending at residue 1,561 from mouse constructs ending at 1,537. The negative cisplatin result applies to the tested screening context. The annotation reviewer did not inspect figure pixels or the entire supplemental experiment set. |
| PMID:27185855(https://doi.org/10.1084/jem.20151183), original | Results Figures 4, 5, 9 and 10, plus cloning/localization/damage Methods. Human 1,561-residue Hebo is nuclear and rescues patient-cell phleomycin sensitivity; expressed 712-residue protein is mostly cytoplasmic and does not rescue in that assay. Forced nuclear short protein also fails damage recruitment. Mild mitomycin-C effects and chromosome abnormalities coexist with non-classic Fanconi readouts. These results do not establish absence of all interstrand-crosslink repair roles; mitochondrial function was explicitly unaddressed. Normal signal-joint fidelity and small homologous-recombination effects are condition-specific and do not erase later evidence for partly redundant end-joining participation. No paper figure pixels were inspected. |
| PMID:41741470(https://doi.org/10.1038/s41467-026-69843-w), original | Figure 5 Results/caption, operative purification/DNA-melting Methods and actual Supplementary Table 3 protein sequences. The purified WT sequence is exactly the 1,550-residue Q5T890-1 product; the engineered mutant differs only at K154A, equivalent to K165 in the displayed product. With magnesium and RPA or mtSSB, ATP enhances melting of short four-nucleotide-overhang substrates compared with AMP-PNP/no ATP. Residual melting means the observed activity is not absolutely ATP-dependent. Blunt/Y substrates and tested conditions delimit this result. The annotation reviewer read text/captions and the sequence proof; an independent reviewer inspected Figure 5. Other supplementary experiments are not claimed read. |
| PMID:24507776(https://doi.org/10.1016/j.ajhg.2014.01.007), indexed original | The separately recovered indexed original includes siRNA/viability/ROS and Subcellular Fractionation Methods, Results on the historical 712-residue GFP construct and genotoxic-stress trafficking, and Figures 2–4 captions. Figure 2A supports short-construct cytoplasmic/nuclear localization. Three-hour MMC/Irofulven exposure, differential extraction with compartment controls, mitochondria-enriched extracts and MitoTracker colocalization support conditional endogenous mitochondrial/nuclear trafficking. Endogenous Abcam-antibody and siRNA experiments are not uniquely resolved to every current product. The ROS-dependent response Results describe NAC inhibition of trafficking and cytotoxicity (Figure 5C/D and Figure S7 references); these figure pixels and supplementary experiments were not inspected. The gamma-H2AX/NAC observation is explicitly data not shown. The complete main file was not recovered; the incomplete normal cache remains unchanged. |
| PMID:20873783(https://doi.org/10.1021/pr100562w), HAL author manuscript and figures, published supplement | The author manuscript Results pages 6 and 9, Methods pages 27–28, Table 1 page 51 and actual Figure 2/4B pixels support bead-immobilized GST-RAD26L from bacterial extracts binding purified full-length His-NEK6 and phosphorylation by wild-type NEK6 but not S206A, with GST and bacterial-lysate controls. Its historical A4D997.3 reagent is residues 1–531; the published supplement Table S1 independently records the fetal-brain screen clones and yeast controls. This author version differs from the indexed final abstract/figure set and does not recover the final target-specific centrosomal colocalization experiment. Omission from this version is not evidence against that curated location. Table S2 predicts sites; it does not negate the observed kinase experiment. No physiological phosphorylation site, equivalence of all current products or native stable assembly is established by these bacterial preparations. |
| PMID:32296183(https://doi.org/10.1038/s41586-020-2188-x), original, HuRI table | Main screen/retest/sequence-confirmation and single-isoform-limit paragraphs, plus six exact PSI method records representing two pairs. CYREN Q9BWK5-1/ORF 20 pairs with target ORF 8144 in screens 1 and 3; TCP11L1 Q9NUJ3/ORF 71146 pairs with ORF 8144 in screen 3. The records do not represent six independent biological interactions. Pair-specific mammalian orthogonal validation and the exact ORF DNA sequence were not recovered. |
The PSI target is H7BXQ9, ENST00000320486.6/ENSP00000320939.5/ENSG00000182150.15, human ORFeome 8144. Official UniProt marks the sequence-source record deleted. UniParc UPI0001610CA1 preserves a 522-residue sequence beginning with X. The remaining 521 residues exactly match displayed Q5T890-4 residues 1041–1561. This sequence projection links the table to the ERCC6L2 C terminus; it is not independent sequencing of clone 8144 or proof that the clone represents a present-day natural product. CYREN also has independent direct human co-IP evidence. TCP11L1 remains a screen-associated partner without a further functional assignment.
The actual PTHR45629 tree places target leaf PTN002473830 below PTN000085120. The interstrand-crosslink repair IBD has self-donor Q5T890, which is legitimate descendant experimental grounding. The target MSA exactly matches all 1,561 residues of Q5T890-4, with no target-path NOT. The unresolved assessment concerns the particular role in crosslink removal/restoration despite newly recovered MMC-sensitivity experiments, not donor count, self-WITH, original-access failure or phylogenetic placement.
ARBA00093742 requires both FunFams 3.40.50.10810:FF:000019 and 3.40.50.300:FF:000821 plus Eukaryota; the actual target satisfies all three conditions. Independent end-joining literature supports the output. InterPro IPR000330 matches the target SNF2 domain; the nucleotide-dependent behavior of purified human Q5T890-1 independently supports ATP binding.
The official HPA record supports nucleoplasm and cytosol. HPA021260 antigen matches residues 1244–1332 of the displayed product and 1233–1321 of the 1,550-residue product; it is absent from the 712-residue product. A-431 and U-251MG show both locations; western validation notes additional bands. HPA022422 matches residues 730–809 and 719–798 of the two long products, respectively, and supports nucleoplasm in Rh30/U2OS. Two selected original HPA images were inspected. These data do not distinguish the two long products or establish the short product's location. The independent GFP-short-protein experiment establishes its broader cytoplasmic distribution. Recovered 2014 endogenous fractionation now independently identifies a cytosolic compartment, so the cytosol refinement does not require transferring the HPA long-product signal to the short construct.
The cytoplasm refinement uses current GO:0005829 cytosol and its typed ancestor route. Current GO:0032991 protein-containing complex requires a stable, jointly functioning assembly; the recovered biochemical RAD26L–NEK6 binding/phosphorylation experiments do not independently settle that scope, and the final-version cellular context remains incompletely recovered. GO:0036297 interstrand cross-link repair describes crosslink removal/restoration, whose experimental and product-specific support remains unresolved.
Bounded official searches for PCNA binding, proliferating-cell-nuclear-antigen binding and DNA sliding-clamp binding did not identify an appropriate specific molecular-function replacement. This does not establish that every possible ontology wording was exhausted. The positive aPIP interaction is retained as non-core. Complex membership, PCNA activation or polymerase chemistry would assert different functions. The kinase-specific NEK6 interaction already has an appropriate term and remains non-core because the wider physiological role is not established by the available experiment.
Operative original Methods and Results have now been recovered separately for PMID:24507776, with the incomplete-file and figure limits stated above. For PMID:20873783, the author-deposited main manuscript and biochemical figure panels are available; the final publisher main version and its target-specific centrosome experiment remain unresolved. The reference-level unavailable flag refers to that final version, while the reference review documents the recovered author version. Normal publication bytes and their availability metadata remain unchanged. PMID:40913148 was genuinely fetched as an additive normal reference; it does not replace any prior cache.
One core combines a bounded DNA-unwinding activity with the established contribution to classical nonhomologous end joining. The biochemical and cellular evidence are complementary; purified end joining was not reconstituted, and the exact coupling of the DNA-melting reaction to the joining machinery remains unresolved.
PMID:41741470(https://doi.org/10.1038/s41467-026-69843-w), original: Results Figure 5 and Methods for protein purification, DNA melting and electrophoretic mobility-shift assays establish short-overhang duplex melting by purified human ERCC6L2. Supplementary Table 3's WT sequence exactly matches the 1,550-residue Q5T890-1 product; the purified mutant changes K154 to alanine. This is distinct from the cellular K154R mutant. Figure 5 pixels were independently inspected during core synthesis. ATP enhances melting relative to AMP-PNP and no ATP, and the K154A mutant loses that enhancement while retaining DNA binding. These controls support an ATP-hydrolysis-coupled component by inference; the inspected experiment does not separately quantify phosphate or ADP release. Residual melting occurs without hydrolyzable ATP. RPA or mtSSB is required under the tested conditions. The assay uses four-nucleotide 5′ or 3′ overhang substrates and does not establish general unwinding of all duplex structures, a directional motor rate, or exonucleolytic resection. The protein sequence proof was read; no additional supplementary experiments are claimed inspected.
PMID:32355287(https://doi.org/10.1038/s41422-020-0328-3), original: Results Figures 3–5 and cell-line, cloning, laser and co-immunoprecipitation Methods establish damage-site recruitment, association with end-joining factors and a contribution to repair-complex loading. Human U2OS and HEK293T experiments are distinct from mouse B-cell class-switch and XLF-redundant V(D)J joining experiments and mouse MEF ligase recruitment. The human construct ending at residue 1,561 and mouse construct ending at residue 1,537 are not the purified 1,550-residue biochemical reagent. DNA-independent co-immunoprecipitation does not establish every pair as a purified binary interaction. The direct role is repair-factor and DNA-end handling, not DNA-ligase chemistry. Text and captions were independently read; no additional figure pixels or complete supplementary experiment set were inspected for this synthesis.
PMID:27185855(https://doi.org/10.1084/jem.20151183), original: Results Figures 4–5 and 9–10, with cDNA cloning, GFP/forced-NLS, microirradiation and repair-assay Methods, distinguish the 1,561-residue nuclear product from the mostly cytoplasmic 712-residue product. Only the long product rescues patient-cell phleomycin sensitivity in the tested assay; forced nuclear localization does not restore the short product's damage recruitment. The condition-specific V(D)J precision and homologous-recombination results do not exclude the later partly redundant end-joining contribution. These observations support the description's product boundaries, not automatic extension of the purified 1,550-residue activity to every product. Text and captions were read; no figure pixels were inspected here.
PMID:37014751(https://doi.org/10.1016/j.celrep.2023.112329), original: Figure 1 centromere localization/DNA-binding-fragment experiments, Figure 4 PCNA-binding experiments and the Discussion limitations support the description's centromere and PCNA statements. Independent core reading included centromere Results, DNA-binding-fragment and end-processing observations, and the authors' limitation on interpreting PCNA-mutant function beyond localization. The accepted annotation packet supplies the detailed PCNA peptide/pull-down/ITC/crystal scope. These data do not establish a separate catalytic centromere activity or a general chromatin-remodeling mechanism. No new figure or supplement inspection is claimed here.
Current GO:0003678 DNA helicase activity requires ATP-hydrolysis-driven DNA unwinding. It is applied here to the supported ATP-dependent component of the measured reaction, with the assay limitation stated in the core itself. It does not describe the entire residual ATP-independent melting signal. Current GO:0097680 classical nonhomologous end joining captures the established end-joining contribution; the description retains the unresolved mechanistic link and the broader cellular effects observed in 2026. Its broader DNA-repair ancestors are omitted to avoid redundancy.
GO:0000729 DNA double-strand break processing specifically describes 5′→3′ exonucleolytic resection and is not assigned to duplex melting. No separate ATP-hydrolysis, translocase, chromatin-remodeler, centromere-catalysis or PCNA-activation core is proposed. Nuclear/centromeric localization remains represented by the accepted annotations and biological description; no location field silently equates the purified 1,550-residue reagent with the separately localized 1,561-residue construct.
PMID:37014751(https://doi.org/10.1016/j.celrep.2023.112329), Results Figure 5A–G and phleomycin/native-BrdU Methods, supports increased pRPA and exposed ssDNA after ERCC6L2 loss in human U2OS cells, without a corresponding increase in gamma-H2AX. CtIP, BLM or BRCA1 depletion reduces that excess processing. The PCNA aPIP region at approximately 790–811 is in the middle of the tested 1,561-residue construct, separate from the C-terminal recruitment/condensate context.
PMID:41741470(https://doi.org/10.1038/s41467-026-69843-w), Results Figures 3–4 with SMART, AsiSI, ChIP and knockdown Methods, adds measured resection tracks, break-associated ssDNA and MRE11 spreading. NBN loss and MRE11/CtIP perturbations define the reported pathway relationships. Unchanged recruitment measurements are distinguished from increased spreading. These Results and Methods were read as text/captions; no additional Figure 3/4 pixels or supplementary experiments were inspected during this feedback assessment.
PMID:40913148(https://doi.org/10.1038/s41556-025-01760-4), original, provides a contrasting prolonged-ATM-inhibition context. Actual official Nature Results and Figures 3–7 captions, together with cell-line, construct, SMART, AsiSI, fractionation, co-immunoprecipitation, decay/ubiquitination and FRAP Methods, were read. In human U2OS cells under 24-hour ATM inhibition, ERCC6L2 loss lowers resection, while condensate-associated CtIP stabilization, RNF138 depletion and complementation support a regulatory contribution. The study declares a 1,561-residue full-length protein, ER-N 1–712 and ER-C 713–1,561, including an NLS-FUS chimera; exact clone sequence equivalence was not independently established. These are cellular experiments, not purified resection or helicase reconstitution. Figure pixels, movies and the supplementary experiment set were not inspected.
Current GO:1903775 regulation of DNA double-strand break processing accommodates this context-dependent direction. Its typed ancestry is regulation of DNA metabolism, and it regulates GO:0000729; the replacement of broad damage response is a lateral functional correction. ERCC6L2 is not assigned exonucleolytic resection chemistry. The regulation annotation remains separate from the catalytic core because coupling of the C-terminal CtIP mechanism to purified helicase activity has not been demonstrated.
One new molecular-function annotation records the measured ATP-dependent component of short-overhang unwinding specifically for Q5T890-1. It is supported by PMID:41741470 and the exact protein-sequence proof already described above, and is absent from the 24 imported assertions. It does not add a process annotation or extend the activity to other products. Current GO:0036121 double-stranded DNA helicase activity additionally specifies dsDNA-dependent ATP hydrolysis; that specificity is not independently resolved by the inspected assay. The bounded GO:0003678 assertion follows the evidence, rather than a desire to eliminate a validation warning.
The project curation instructions explicitly retain supported generic binding as non-core when no supported functional replacement is available and prohibit new publication quotations. The four supported generic interactions remain non-core under that instruction; no speculative adapter function is manufactured. Interstrand-crosslink repair remains uncertain after recovery of the original MMC experiments because the particular repair step remains unresolved, not because every process assertion requires an isolated biochemical assay.