This is an authored source and locator guide. No new publication quotations were added. Original enzyme, interaction and localization evidence was assessed together with database provenance. Normal publication caches remain unchanged; private recoveries supplement their access limits. Provider research failed during intake and no provider report was manufactured. For PMID:22833457, original PDF text was readable through web extraction; the separate 919-byte local download is an HTTP403 failure body, not a valid PDF. Actual returned web text is preserved separately from that failed request.
PMID:10329163. The normal cache is abstract-only. The author-uploaded original was recovered through indexed web access despite direct HTTP 403 responses. Read original Results on expression and physical properties, ATP:pyrophosphate exchange, tRNA aminoacylation/Table 1, and Methods on cloning, protein purification, exchange and aminoacylation. The private captured read spans include original lines 455–928, 1108–1373 and 1836–2302. Recombinant human enzyme directly catalyzes charging. The bacterial transcript and bovine mitochondrial tRNA substrates do not establish cellular localization of the expressed human protein. The targeting/cleavage boundary was predicted; no exact natural processing site or isoform is asserted. Functional tRNA recognition is demonstrated by charging/steady-state kinetics, not an equilibrium binding measurement.
PMID:19549855. Read the complete cached original main text, including Results, Discussion and Methods (original). Human mitochondrial and cytosolic PheRS were separately tested with bacterial/yeast tRNA substrates. Human mitochondrial PheRS directly charges phenylalanine and also meta-tyrosine in vitro. Cytosolic/bacterial editing findings are not transferred to FARS2. Figure pixels were not inspected.
PMID:22833457. The normal cache is abstract-only. Recovered the original nine-page article PDF and author-uploaded text. Read functional analysis of recombinant mtPheRS variants, table-linked kinetic results, and aminoacylation/ATP:pyrophosphate exchange Methods (PDF pp.5–7, text lines 241–280 and 439–493; author text Results lines 684–779). Wild-type human enzyme provides the positive activity comparator. Changes in ATP, amino acid and tRNA kinetics are distinguished from structural predictions. Patient fibroblasts did not reproduce the reported respiratory/translation defect; disease-associated necessity does not itself license additional GO processes.
Official QuickGO GO:0006432 marks the imported term obsolete and explains that it duplicates GO:0004826, with biological process coverage in GO:0006418. The GO:0008033 definition concerns pre-tRNA maturation before aminoacylation. Consequently, the phenylalanyl-tRNA aminoacylation and tRNA-processing assertions are refined to GO:0006418; original imported IDs and labels remain exact.
Every one of the 81 protein-binding annotations has an exact, PMID-specific IntAct record with O95363 and the imported partner accession, including any imported partner suffix. The row-level receipt table contains target and partner accessions, method, record IDs and author-release comparison. These positive records are retained as non-core under the project's explicit binding convention. Association supplies no evidence that FARS2 executes a partner's catalytic or signaling role, and none was assigned.
PMID:25416956, HI-II-14. Read original Results and experimental procedures and Extended Experimental Procedures section 3: ORF construction, yeast strains, autoactivator removal, screening, sequence verification, and quadruplicate pairwise retesting. The original pipeline retains pairs positive in at least three of four tests with sequence confirmation. All 22 HI-II-14 annotations have exact human IntAct records. The archived author PSI release provides additional target rows, but its historical partner identifiers do not always equal today's accessions. In particular, partners A1A4E9 and O75478 lack an exact/base-accession author join in that release; the current PMID-specific IntAct records remain positive. Neither missing archival mapping nor an isoform suffix change is treated as disproving the interaction. No individual orthogonal-validation claim is extrapolated from global validation rates.
PMID:32296183, HuRI. Read original generation/retesting Results, Extended Data Figure 1 caption and Discussion. Nine screens using three Y2H versions were followed by quadruplicate pairwise verification and sequence confirmation. All 55 HuRI annotations have exact human IntAct records. The author PSI release is separately inspected; partners A8MQ03, Q9UFD9 and Q9ULR0-1 lack an exact/base-accession join, and other partner isoform identifiers differ. Current curated source records support these contexts; this review does not reconstruct or guess an isoform remapping. Y2H establishes interaction potential under the assay conditions, not an endogenous cellular function or compartment. The detailed original supplement for HuRI was not separately read; the original main explicitly documents the operative verification procedure.
PMID:31515488. Read original Results and Methods, particularly “Profiling disrupted protein interactions through Y2H” and orthogonal PCA retesting. Actual Supplementary Data 2 (41467_2019_11959_MOESM6_ESM.xlsx) row 797 records APPL1/Q9UKG1 E67G and row 3201 CALCOCO2/Q13137 E222K; Supplementary Data 3 (...MOESM7...) row 67 records APPL1 E643Q. Each has interactor Entrez GeneID 10667 (FARS2) and Disruption 0. These are mutations in the partners. The Methods explicitly use positive wild-type interactions as the comparator and require reproducible growth reduction to call disruption. Exact IntAct records EBI-22141157 and EBI-22142192 corroborate the human pairs. No FARS2 variant, individual PCA validation or partner-derived pathway role is inferred.
PMID:28514442 and PMID:33961781, BioPlex. Read original clone/culture, AP-MS, scoring and network-assembly Methods (2017 original, 2021 original). Anti-tag purification and mass spectrometry report association that can be indirect. The exact FARS2–GAMT/Q14353 records are EBI-21814498 (2017) and the corresponding 2021 record in FARS2-binding-records.yaml. The author BioPlex 3.0 293T release contains the exact O95363–Q14353 edge at line 63913, with pInt 0.997674438312471; no matching edge was found in the inspected HCT116 release. That absence is release-specific, not a demonstrated biological negative. BioPlex 3.0 reanalyzes older datasets as well as adding experiments; the two papers are not assumed independent replicates. No direct contact, creatine-pathway activity or shared localization is inferred.
PMID:34800366. Read original Results on subtractive/spatial proteomics, TOMM40 importomics and MitoCoP construction, plus cell culture, experimental design, fractionation and statistical classification Methods. Actual author Tables S1–S3 identify the target by FARS2/O95363:
mmc2.xlsx), row 334: inclusion in the 1,134-gene MitoCoP.mmc3.xlsx), row 1554: subtractive Class 1; mean log2 purified/crude mitochondrial ratio 1.1035, four valid ratios.mmc4.xlsx), row 1233: TOMM40-importomics Class 1; mean log2 knockdown/mock −1.6975, p=0.0001085, four valid values.These are actual positive target measurements, beyond inclusion in a summary catalogue. They support the broad mitochondrial HTP annotation. The measured protein group includes O95363/Q5JRF7 with O95363 the majority/simplified identifier; it is not evidence resolving a specific natural FARS2 isoform. No figure pixels were required to interpret these positive numerical tables. Incidental numerical 10667 hits in unrelated protein-group ID columns were excluded from target evidence.
Rat Q6AYQ3 is Fars2. Its mitochondrial location is experimentally supported by PMID:26553276(https://pubmed.ncbi.nlm.nih.gov/26553276/), whereas its matrix statement is curator inference ECO:0000305. Human O95363 transfers both locations from this rat record. These source paths do not establish an experimental rat matrix-localization assay.
The normal PMID:26553276 cache is abstract-only. Publisher full-text/PDF requests returned 403, but an author-uploaded accepted manuscript was read. Original Results lines 308–333 report rat CNS expression and Fars2/COXIV colocalization; lines 348–373 report recombinant human WT/D142Y enzyme assays. A refreshed source response places the Figure 2 rat immunostaining caption, including COXIV staining and omitted-primary-antibody control, at lines 647–668. The responses have different page numbering; they are not a single reconstructed transcript. Figure pixels and supplementary methods were not inspected. Human enzyme assays and rat tissue staining have different species/construct scope.
The matrix annotation and the core's matrix location are retained as curated inference supported by UniProt source tracing and Reactome's explicit O95363 matrix catalyst. Their recurrence across ISS, IEA and TAS records does not increase experimental precision. Human MitoCoP measurements directly support mitochondrion, while its peripheral-membrane class does not identify a submitochondrial compartment. GO:0005737 cytoplasm includes organelles and is compatible with mitochondria; it does not establish cytosolic FARS2 or the cytosolic PheRS complex.
Both normal Reactome records and current ContentService objects were read. R-HSA-379726 contains R-HSA-380203, which connects through catalystActivity 379675 to physicalEntity 379565. That entity is human O95363, carries matrix compartment GO:0005759, and supplies GO:0004826 catalytic activity. The reaction consumes phenylalanine, tRNA(Phe) and ATP and produces charged tRNA, AMP and diphosphate. The inspected target route supports FARS2's own step, without assigning other pathway catalysts. The pathway's general protein-import citation PMID:11121736 was not independently read and supplies no target-specific evidence here.
The actual PANTHER tree and MSA place O95363 at PTN002480027 in PTHR11538:SF41. The 22-node ancestry path passes PTN000165932 and PTN000165999; the ungapped target MSA reproduces the official 451-residue sequence. The PAINT file places Phe ligase, aminoacylation and cytoplasm assertions at PTN000165932, and mitochondrial location at PTN000165999. Their dates are 2023-01-10, 2025-09-02, 2026-05-28 and 2026-05-28, respectively. Function-loss assertions at PTN000711660 and the cytosolic PheRS complex at PTN000165935 are off the target path. Target self-WITH is legitimate descendant experimental grounding; donor number is not an evidence-strength test. The aminoacylation assertion remains biologically supported while its obsolete term is refined to GO:0006418.
Actual InterPro O95363 matches include IPR002319 at residues 76–198 and 242–343, and IPR004530 at 225–446. Substrate-binding and catalytic mappings agree with human biochemical evidence. RHEA:19413 and EC 6.1.1.20 specify phenylalanine-tRNA ligation. The current ARBA00026908 rule has a broad ligase consequent, but no complete matching target condition set was independently reproduced: its superficially corresponding two-FunFam clause contains FF000053 where the target lists FF000041. Independent human biochemistry and the actual InterPro match support the broad activity; the unresolved rule clause does not demonstrate loss of function.
The complete official GO_REF:0000002, 0000024, 0000033, 0000044 and 0000120 method descriptions were checked against their target records. Method titles supply provenance, not gene-specific experiments; merged electronic routes are not independent experimental replication.
UniProt displays a 451-residue precursor but specifies no exact mature cleavage boundary or curated alternative products. RefSeq and Ensembl transcript/protein models, predicted recombinant boundaries and Reactome's FARS2(1-?) entity do not establish additional natural products. No synthetic O95363-1, guessed isoform or natural processing coordinate is supplied.
The core is ATP-dependent phenylalanyl-tRNA synthesis. tRNA and ATP binding are catalytic substrate interactions. The cytosolic enzyme's editing function is not transferred to FARS2. Meta-tyrosine charging remains an observed in-vitro substrate property, and disease consequences remain biological context. No additional process or activity is inferred from interaction partners or necessity phenotypes.
The broad nucleotide-binding, aminoacyl-tRNA-ligase and tRNA-charging assertions are true but subsumed by retained ATP binding (GO:0005524), phenylalanine-tRNA ligase activity (GO:0004826) and tRNA aminoacylation for protein translation (GO:0006418). Both cytoplasm assertions are likewise subsumed by the positively supported mitochondrion annotation. These five decisions use TOO_GENERAL; the original source objects, PAINT placement and unresolved ARBA eligibility remain intact. The independently informative mitochondrion rows remain accepted. The core now lists mitochondrion alongside matrix and explicitly identifies matrix placement as a curated inference.
The binding records are now authored YAML, with each IntAct method represented as a list item. All 81 records and every original field are retained, including empty author-release locators, exact partner suffixes and accession-limit caveats.
In PMID:19549855, Results under “Discrimination Between Cognate and Noncognate Substrates” establish alternative-substrate charging by the human mitochondrial enzyme. Methods under “tRNA Aminoacylation” specify bacterial or yeast tRNA substrates. The Discussion proposes a mitochondrial contribution to meta-tyrosine incorporation. These experiments do not measure FARS2-dependent incorporation into mitochondrial proteins in living human cells. The knowledge gap and proposed experiment are bounded to that distinction; they add no physiological activity, editing function or oxidative-stress annotation. This is not a claim that a comprehensive later-literature search found no relevant study.