FANCM combines ATP-driven remodeling of branched nucleic acids with recruitment of DNA-repair machinery to chromatin. These activities are complementary. MHF-assisted fork remodeling and FANCM:FAAP24 hybrid remodeling were tested in different protein preparations; a result from one preparation is not automatically assigned to the other. The three annotated natural products are also distinct from engineered fragments and patient alleles.
The review retains historical quotations and findings for provenance. Some older conclusions in the notes and reference histories are superseded by the dated reassessment: RNA-containing substrates are positively supported, normal RAD51 foci do not exclude downstream SDSA participation, and chromatin enrichment in tested cells does not exclude all other locations. New judgments appear in annotation reviews and the current biological summary.
| Public source | Evidence used and limits |
|---|---|
| PMID 20347428, original article | Selected main Results, Methods pointers and captions cover native human assemblies, recombinant full-length FANCM–MHF fork reversal, engineered FANCM residues 1–754 binding, ICL recruitment, chicken DT40 mitotic SCE and fission-yeast blocked-fork gene conversion. The ordinary publication cache is abstract-only; the original text was recovered separately. Detailed supplementary procedures and figure pixels were not inspected. The human meiotic cleavage/rejoining assertion remains unresolved. |
| PMID 20347429, original article | Selected main Methods, Results and captions cover human nuclear-complex purification, FANCM-dependent MHF association with FA-core proteins, HeLa chromatin fractionation, MHF1 perturbation, purified MHF1–FANCM pull-down, ATPase and movable-fork branch migration. MHF stimulates branch migration without increasing measured ATPase potency. Several detailed procedures are referred to supplements, which were not read. The fractionation result is not a universal localization exclusion. |
| PMID 26593718, original article | The sections on FANCM-mediated R-loop resolution and their biochemical Methods show ATP-dependent DNA:RNA-flap displacement by FANCM:FAAP24, with K117R, ATP analog and homology controls. Human cellular perturbations and chicken DT40 knock-ins are separate evidence. Selected text was read; figure pixels and supplements were not inspected. |
| PMID 31138795, original article | Telomeric R-loop Results and protein-purification/displacement Methods show that recombinant FANCM:FAAP24 releases RNA from plasmid R-loops in an ATP-dependent assay. This is displacement, not RNA degradation. U2OS wild-type/K117R complementation and BLM/RNase H perturbations provide a distinct ALT-cell context. Selected text was read; no natural-product equivalence or complete supplementary audit is claimed. |
| PMID 24207054, original article | Selected operative Results and captions support replication traverse of ICLs with motor/DNA-binding dependence. Traverse is not itself lesion excision; the broader ICL-repair assignment also uses independent FA-pathway evidence. |
| PMID 29231814, original article | Selected human tissue, patient-cell and complementation Methods/Results/captions support nuclear/pachytene localization and defective FANCD2 monoubiquitination improved by wild-type FANCM. The p.Gln1701* product is a partially functional patient allele, not a complete null or one of the annotated natural alternative products. Pachytene localization does not demonstrate meiotic intermediate cleavage/rejoining. |
| PMID 23333308, original article | Selected HCT116 knockout Methods/Results distinguish checkpoint outcomes from acute depletion and discuss possible adaptation. Together with the non-null patient results, this limits the inherited broad checkpoint proposal; it does not establish that FANCM has no checkpoint-related function. |
| PMID 32769987, original article | Selected main Methods, Results and Figure 2 captions support MCM2–FANCM association by sequential chromatin IP/PLA. This is complex/proximity evidence, without an inference of isolated binary contact. |
Broad nucleus and ATPase assertions with inaccessible original assays are retained where the curator's interpretation agrees with independently inspected human evidence. The original operative Methods/Results were not recovered for PMID 16116422, PMID 17289582 or PMID 18206976. The available PMID 19423727 record contains abstract/Introduction/Discussion text rather than a verified complete operative body. Its inherited quotations are preserved, without claiming that inaccessible experiments were reread. PMID 18995830 supplies positive checkpoint/translocase information at abstract level; the original operative body was not recovered.
The verified PTHR14025 v19 tree places human leaf PTN002493637 below PTN000361481. The target MSA sequence exactly matches displayed 2048-residue Q8IYD8. Five term-specific ancestral IBDs match the source annotations. A short donor list is not a defect, and human FANCM appearing among its own ICL-repair descendant sources is legitimate experimental grounding.
| Public source | Donor experiment and boundary |
|---|---|
| PMID 18851838 | Selected Fml1 Results/Methods examine an engineered 603-residue truncation, mobile-junction processing, D-loop dissociation and fork reversal. Substrate-dependent limited unwinding is positive donor activity, not a human construct result. |
| PMID 22723423 | The complete recovered main BioC text and captions were read. Fml1 truncation/mutant D-loop assays and meiotic noncrossover genetics support donor function; supplementary procedures and figure pixels were not inspected. |
| PMID 28977643 | Selected centromeric recombination Results, captions and Discussion support Fml1/Mhf crossover suppression. The SDSA explanation is a mechanistic model, not a complete purified-human pathway assay. |
| PMID 25205745 | Fly P{wa} gap repair directly distinguishes completed SDSA products and finds reduced SDSA in Fancm mutants. The shorter repair-synthesis-tract result concerns Blm, not Fancm. Selected Methods/Results/Discussion were read. |
| PMID 27466228 | Recovered full-text HTML and operative Methods/Results describe an engineered fly protein lacking 840 C-terminal residues, with ATP/Mg, K84M and substrate-polarity controls supporting limited 3′–5′ unwinding. Catalytic and noncatalytic in-vivo roles are separable; the human protein was not assayed. |
| PMID 15634678 | The verified abstract supports Mph1 ATPase and RPA-enhanced 3′–5′ DNA helicase activity. Full Methods/Results were inaccessible; unreported construct details are not inferred. |
DNA translocase activity is the precise human motor description. This refinement preserves positive donor chemistry and ancestral placement. Four-way-junction binding and ATP-dependent junction processing remain accepted. The actual GO:0009378 definition is compatible with the observed branched substrate output; it does not require a claim of unrestricted linear-duplex unwinding. GO:0045003 names noncrossover SDSA, whereas GO:0035825 describes equal-material exchange. The captured QuickGO is_a/part_of graph places SDSA below GO:0000724 (double-strand break repair via homologous recombination), but has no path from SDSA to GO:0035825. This does not establish biological disjointness: reciprocal recombination is a narrow synonym and crossover a related synonym, so GO:0035825 is not restated as crossover-only. Positive mechanistic evidence motivates the SDSA refinement.
UniProt Q8IYD8 supplies the exact source protein and natural-product records. The 669-residue product has a substitution at residue 669 and lacks residues 670–2048. The 2022-residue product lacks residues 228–253 in the motor region. Engineered FANCM 1–754 and the patient p.Gln1701* product are different entities. Neither a shared gene symbol nor an overlapping antibody epitope establishes equivalent activity.
HPA FANCM subcellular evidence assigns supported nucleoplasm localization using HPA055144. Actual A-431 and U-2 OS image fields were inspected; PC-3 was assessed from metadata only. The antigen maps to displayed residues 1802–1886, absent from the 669-residue product, and does not distinguish the two longer products.
All 13 cited Reactome events were traced to FANCM's own compartment-bearing entity rather than assigned the enclosing complex's compartment by assumption. Twelve nuclear events include Q8IYD8 residues 1–2048 in nucleoplasm. The cytosolic PKR-regulation event R-HSA-9835411 reaches the cytosolic FANCM:FAAP24 entity. That contextual location is retained as non-core. The event's FANCC/FA-pathway perturbation evidence does not demonstrate isolated FANCM inhibition of PKR. Traces record representative routes to each reached entity, not every possible duplicate graph route.
InterPro mappings are retained at their supported scope: IPR006166 for the ERCC4-like DNA-binding region, IPR011545/IPR044749 for N-terminal nucleotide-binding regions, and IPR039686 for the FANCM/Mph1 insertion. A fold annotation does not independently establish nuclease chemistry. Rhea 13065 supports ATP hydrolysis; the EC RNA-helicase mapping is refined without denying demonstrated RNA:DNA remodeling.
Seven supported generic protein-binding rows remain non-core under the project policy. FAAP24 and CENPS have native/purified-complex evidence; TRIM27 and EPN2-2 have target two-hybrid records; MCM2 has chromatin-complex/proximity evidence. The larger screens use actual target records from PMID 25416956, PMID 32296183, PMID 33961781 and PMID 40205054. In the last study, the author-linked NDEx network 95bc75d5-d1d1-11ee-8a40-005056ae23aa contains FAAP24–FANCM edge 565778. An initial spreadsheet download was recaptcha HTML and was not treated as a workbook. Association alone is not converted into catalytic, partner-class or process activity.
No new quotations were added. All inherited quotation positions, text and reference associations remain intact. Full-text availability flags describe a cache and are not proof that every section, figure or supplement was read. Selected original-body scope is stated above and in each reference review. The identified 2022 R-loop follow-up, PMID 36476850, remained inaccessible in full text and is not required evidence for the current judgments.
The historical GO:0035825 / IEA / GO_REF:0000117 row records ARBA:ARBA00027491 in the saved source TSV. The official current rule remains available and outputs GO:0035825 through 16 condition sets. The earlier reference note that the precise historical rule could not be recovered is corrected by that source record. Its historical target-qualifying condition and the cause of the target row's absence from fresh GOA remain unresolved. The row is retained as retired history under an explicit manual source-lifecycle decision; its MODIFY interpretation and exact quotation remain. The independent, positive SDSA IBA stays active. Rule deletion, biological absence and automatic retirement are not inferred.
The complete inherited Affinage report was read for discovery. Its motor, hybrid-remodeling, FA recruitment and donor-recombination themes overlap independently inspected primary sources. Other leads include MM1/MM2–BTR bridging, PCNA/AND-1 recruitment, ATR/PLK1 regulation, newer structures, and synthetic-lethal or PARP-inhibitor phenotypes. Report-specific claims in those leads were not independently verified in this integration and were not imported as new core functions or GO assertions. The report mixes species, constructs and broad checkpoint language; its self-evaluation scores and proposed GO/Reactome grounding are not evidence validation. The machine-fetched report remains unchanged and is not added as annotation support solely to suppress a validator advisory.
The structured core now includes R-loop processing, GO:0062176, supported by the existing human FANCM:FAAP24 hybrid-flap and telomeric R-loop experiments PMID:26593718(https://pubmed.ncbi.nlm.nih.gov/26593718/), Figure 6 and Biochemical Analysis; PMID:31138795(https://pubmed.ncbi.nlm.nih.gov/31138795/), Figure 7b and In vitro R-loop resolution assays. A reference-only IC/NEW proposal records that process. The enzyme performs the disassembly step; the claim is not based solely on a knockout phenotype. Hybrid binding is not added as a separate molecular function from these catalytic assays.
Chromatin-protein adaptor activity, GO:0140463 expresses the established FA core recruitment role more precisely than chromatin binding alone. The seven pairwise association contexts retain their individual decisions: MHF and FAAP24 aid FANCM DNA engagement, whereas the recruitment core concerns FA core cargo. The description restores supported family, phosphorylation, ovarian-expression and ovarian-insufficiency biology; natural-product functional equivalence remains an experimental question.
Shared Reactome access scope, consolidated from twelve repeated reference notes: the prior traversal records representative routes rather than every duplicate route; unresolved off-target branches remain in the evidence packet. Cited primary papers were not all independently read. These model records are not substituted for original experiments. Event-specific judgments and the distinct cytosolic event caveat remain in each reference review.
The captured human DDX21 and DDX23 records assign GO:0062176 with involved_in and IDA. The budding-yeast SEN1 record uses acts_upstream_of_or_within and IMP; it is a same-term convention check, not direct participation evidence. The two human remodeling proteins provide the direct-participation comparators. FANCM participation rests on its own purified-protein disassembly experiments, not on these comparator annotations.