Cathepsin D is an aspartic endopeptidase whose central performed role is endolysosomal proteolysis. The official human P07339 record has one 412-aa precursor and no alternative-product listing. Signal 1–20 and propeptide 21–64 precede the active region 65–412; current light/heavy chains are 65–162 and 169–412. These are proteolytic products, not splice isoforms or evidence of precursor homodimerization. The native/pepstatin structural abstract supports the human aspartic mechanism, while the human mutant study shows that a stable single chain remains active in rodent cells [PMID:8393577; PMID:17188016].
The two Reactome cysteine-type activity rows are removed for an actual mechanistic contradiction. Both resolve through catalyst 2130638 and CandidateSet 2130682 to CTSD member 1181223; human CTSD uses Asp97/Asp295, not the cysteine nucleophile specified by GO:0004197. Positive lysosomal-lumen location and broader modeled antigen-processing participation remain separate judgments. Event 2130336 resolves to LGMN and is not substituted as CTSD-specific evidence. Likewise, a serine-type label attached to the mixed endostatin catalyst is not assigned to CTSD. Reaction 9023846 places CTSD as the expressed product, not as a transcriptional regulator.
The complete 16997486 abstract reports cleavage of both lipid-free and lipidated human apoE. The cathepsin reagent species is unstated, so the activity is retained without inventing purified-human-enzyme Methods. Amyloid-beta(1–40)/(1–42) binding/cleavage in17112520 is peptide-relative numbering; the literal MINT feature is retained but does not identify the corresponding APP precursor coordinates. The GRN original and supplementary experiments use human tagged CTSD 19–end and human progranulin/granulin constructs in human or mouse cell contexts. All seven processed granulin pairs remain distinct. Co-immunoprecipitation from cells or media is not relabeled purified binary binding [PMID:16997486; PMID:17112520; PMID:28493053].
The apoptotic context is upstream regulation. Actual human-fibroblast experiments show cytochrome-c redistribution before caspase activity after active-enzyme microinjection; later endogenous lysosomal leakage is distinct. Sigma C8696 reagent species is unspecified. The work does not establish direct BID cleavage or CTSD execution-phase dismantling. GO:0090200 explicitly requires DIRECT positive regulation and is therefore not substituted for the supported broader protein-localization regulation PMID:12107093.
The insulin-source chain is now explicit. The 2016 mouse study measures endosomal enzyme activity kits and receptor distribution/phosphorylation. Its reference 34 is Authier et al. 2002: rat-endosome immunodepletion and bovine purified enzyme directly demonstrate insulin cleavage; matching human digestion is stated as data not shown. This supports contextual insulin catabolism and lumenal location. Neither the kit signal nor receptor phosphorylation resolves return to a signal-ready plasma membrane, so the recycling assertion remains undecided [PMID:27703010; PMID:11779865].
All 91 genuine seeded objects remain literal, including qualifiers, donors and processed-product identifiers. There are 30 IPI objects. The full current interaction capture contains 124 IntAct plus 10 MINT records, all 42 fields retained in the separate audit. Exact source/partner joins use 64 Haenig screen records and 2 APP records; 132 participant ROGIDs reconstruct exactly. Every partner name/species is read from its accession-specific official record. In particular, P45880 is displayed VDAC2 isoform 3; P28347-2 is TEAD1; Q15007-2 is 151-aa WTAP; Q9Y6F6-3 is IRAG1 with a 207–803 assayed feature. No partner is assigned by a remembered symbol or an ambiguous alias.
Six screen refinements describe well-defined partner classes while preserving assay limitations: WWP2 E3-ligase binding, TEAD1 Pol II-specific DNA-binding-factor binding, VDAC2 transmembrane-transporter binding, H3.1 histone binding, KIF1A kinesin binding and GPR141 receptor binding. The orphan-receptor and splice-product qualifications remain explicit. Other positive interactions remain non-core where no better established class is supported. This follows the standing project instruction in CLINGEN_MENDELIAN.md, retaining supported generic protein-binding assertions; the validator's generic-binding suggestions do not override that source-specific policy. Binding does not transfer a partner's catalytic activity or establish endogenous colocalization.
The actual PTHR47966 tree has 1286 nodes/769 leaves; target PTN002492008 has an exact 412-aa MSA match. All four literal IBDs lie on its path: catalysis/proteolysis at PTN002492107 and lysosome/extracellular context at PTN008723784. The complete 16 family GAF records and 26 complete donor queries (699 records) are preserved in the audit. Self-grounded human evidence is legitimate. Current exact-term absence does not negate historical or child-term evidence. ARBA00026971 has 2388 branches; only zero-based branch 603 matches both recorded FunFams. Its broad cytoplasm result is compatible with cytoplasmic organelles and is not a cytosol-residency claim.
Genuine ECM supplements were reused byte-exact and CTSD cells independently read. Colon/liver S1 row 205 occurs in both full-data and matrisome views; these are not separate experiments. Myeloma S2 row 29 and S3 row 93/group sheets establish P07339, with differing detection summaries preserved. The breast/omentum workbook contains 11 CTSD measurement rows plus 2 derivative summaries. Human aortic, venous and HUVEC target tables were also inspected. Fraction association does not establish a structural matrix role, transmembrane topology or downstream disease mechanism [PMID:25037231; PMID:28344315; PMID:28675934; PMID:20551380; PMID:27068509; PMID:25204797].
The author urinary-exosome database explicitly lists CTSD/NP_001900, one peptide, reference 2 to 19056867. The separate 23533145 CTSD target supplement was not recovered; files containing HTML were not treated as spreadsheets. Colostrum16502470, plasma 14718574 and pig-ECM 22261194 target tables also remain unresolved. Membrane-specific 17188016 assertions and insulin-receptor recycling remain undecided, without denying the broader positive localization or activity. Source-specific uncertainties explain the two same-term action advisories [PMID:19056867; PMID:23533145].
The compact source evidence index records public identifiers, target locators, product distinctions and access limits. Complete raw responses and reconstruction inputs remain in the separately preserved audit packet. No P07339 GO-CAM activity occurs in the selected immutable index; no new assertion is inferred from that absence.
ROOT performed the genuine isolated gene fetch and publication caching. Default Falcon and its bounded fallback failed terminally with authentication/quota responses; no provider report exists. This document is manual synthesis. Existing publication, GOA, UniProt, GO-CAM and protected source bytes were retained. The two shared caches match their pending owners exactly; 18337546 remains absent from the captured main and requires the recorded shared-source publication qualification. Four additional publications 8393577,27703010,19710420 and11779865 were fetched by the ordinary publication mechanism after path/ownership qualification.
Independent source consultations covered catalytic chemistry/Reactome (ROOT, with Bloc challenge), apoptosis (ROOT), and the insulin-source chain (Bloc). The owner read their complete recommendations and access ledgers and reproduced their read-only checks, while independently reconstructing remaining source scopes. No full-paper access is claimed for an abstract-only or failed retrieval, and repository cache flags remain unchanged even when an authentic indexed original was read separately.
The biological summary and catalytic core now describe the enzyme without source-annotation framing. The broad cytoplasm assertion is retained as a contextual location; the established endosomal and lysosomal lumen terms remain the core locations. The current UniProt mass-spectrometry statement supplies a direct repository anchor for the electronic melanosome mapping, without claiming that the original target table or imaging was read.
The standing CLINGEN_MENDELIAN policy retains supported generic protein binding as KEEP_AS_NON_CORE when no more precise class is established. The six existing positive class refinements are retained after reading the actual definitions and reconstructing the exact partner products. In particular, signaling receptor binding does not require ligand-site occupancy or activation. The NeuroNet original (Haenig et al., PMID:32814053; https://pub.dzne.de/record/153392/files/DZNE-2020-01389.pdf) describes pooled screening followed by pairwise retests in yeast; the three literal records per selected pair are not three independent physiological studies. Selected screening Results, Figure 1 caption and automated Y2H Methods were read; no target-specific mammalian validation or CTSD processing was inferred.
TEAD1-2 retains the TEA DNA-binding region despite its activation-region deletion. VDAC2 isoform 3 is the official displayed channel sequence. WTAP-2 is a regulatory rather than catalytic component. ZDHHC20-4 retains residues 1–314: the original full-length human enzyme study (PMID:29326245; https://pmc.ncbi.nlm.nih.gov/articles/PMC6317078/) maps its resolved catalytic architecture to 5–299 and functionally tests motifs upstream of the changed tail. Row 25 therefore refines to enzyme binding as an explicit class inference, not an isoform-specific enzyme assay. MOGAT3-3 retains Cys265 and membrane segments, but the retrieved evidence does not delimit a complete catalytic unit excluding its replaced tail; its positive generic association remains contextual without an inactivity claim. P09565 is a putative 113-residue IGF2-associated product with no assigned official symbol.
Antigen processing is retained contextually with two additional source anchors. The complete abstract of PMID:1328388 reports purified cathepsin D producing ovalbumin material recognized by MHC-II-restricted T cells on fixed antigen-presenting cells. The original publisher route was unavailable; enzyme species and detailed controls remain unverified. The relevant original mouse Methods, Results/Discussion and Figure 2 caption of PMID:9539769 (https://pmc.ncbi.nlm.nih.gov/articles/PMC22521/) show modest shifts in selected epitope presentation with preserved overall presentation and invariant-chain degradation. These are compatible with a nonessential proteolytic contribution. They do not validate the mixed Reactome set's cysteine chemistry or a CTSD-specific lip10 cleavage step. The normal caches remain abstract-only even though the 1998 original was separately read.
Insulin receptor recycling remains UNDECIDED because the accessible experiments do not resolve the specific return process; the enum is not limited to inaccessible literature. Existing biological source objects, processed-product fields, all prior excerpts and this notes prefix remain exact. The separate TMP audit records actual read boundaries and the ten-row, 30-record, 60-backbone reconstruction. No new CTSD activity or process annotation is proposed.