{"gene":"TG_283780","gene_symbol":"GPI","run_date":"2026-09-11T11:01:00","timeline":{"discoveries":[],"current_model":""},"narrative":{"mechanistic_narrative":"No mechanistic discoveries found in literature.","teleology":[],"mechanism_profile":null,"strain_conflicts":[]},"prefetch_data":{},"corpus_meta":[],"cost":{"stage1":{"model":"claude-sonnet-4-6","input_tokens":28538,"output_tokens":7091,"usd":0.09599,"stage1_stop_reason":"end_turn"},"stage2":{"model":"claude-opus-4-8","input_tokens":14929,"output_tokens":7932,"usd":0.136472,"stage2_stop_reason":"end_turn"},"total_usd":0.232462,"stage1_batch_id":"msgbatch_015xmQKDRqnTh9anE4m7LiD6","stage2_batch_id":"msgbatch_01JLpkFTVhoQBNgqHCKFrVVU","note":"batch pricing = 50% of standard","corrective_rerun":"2026-09-10 wrong-gene rerun (round-4 rules)","corrective_reason":"summary amended","hand_review_patch":"2026-09-11 AK hand-review verdicts (handoff/tools_2026-09-11/patch)","hand_review_verdict":"blank"},"stage1_raw":"```json\n{\n  \"discoveries\": [\n    {\n      \"year\": 1992,\n      \"finding\": \"A cell-free system derived from T. gondii tachyzoite membranes catalyzes GPI biosynthesis via sequential glycosylation of phosphatidylinositol, using GDP-mannose and UDP-GlcNAc as sugar nucleotide donors, with Dol-P-mannose involvement demonstrated by dolichylphosphate stimulation and amphomycin inhibition. The core glycan is assembled similarly to T. brucei but additionally modified by N-acetylgalactosamine.\",\n      \"method\": \"Cell-free biosynthesis assay with radiolabeled sugar nucleotides, chemical/enzymatic tests, chromatographic analysis, dolichylphosphate stimulation/amphomycin inhibition\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — in vitro reconstitution of GPI biosynthesis with multiple orthogonal biochemical methods in a single rigorous study\",\n      \"pmids\": [\"1328232\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1992,\n      \"finding\": \"Four major free glycolipids (GPI precursors) extracted from T. gondii tachyzoites share structural features with the GPI anchor of surface protein P30: a non-acetylated glucosamine-inositol phosphate linkage, PI-PLC and nitrous acid sensitivity, identical HF-dephosphorylated glycan backbone to P30 anchor, and terminal ethanolamine phosphate. These free GPIs are identified as candidate precursors in GPI anchor biosynthesis.\",\n      \"method\": \"Metabolic radiolabeling, enzymatic (PI-PLC, nitrous acid) and chemical sensitivity assays, chromatographic analysis\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — multiple orthogonal biochemical methods establishing GPI precursor identity with rigorous controls\",\n      \"pmids\": [\"1534801\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1993,\n      \"finding\": \"The T. gondii tachyzoite surface glycoprotein gp23 is attached to the parasite membrane by a GPI anchor. The anchor contains a conserved core structure (ethanolamine-PO4-6-Man-Man-Man-GlcN-PI) bearing beta-linked N-acetylgalactosamine residue(s), and is released from the parasite surface by PI-PLC treatment.\",\n      \"method\": \"Metabolic radiolabeling with multiple precursors, PI-PLC and PLA2 digestion, exo/endoglycosidase treatment, immunoprecipitation\",\n      \"journal\": \"Biology of the cell\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Moderate — multiple orthogonal biochemical methods establishing GPI anchor structure of gp23 in a single study\",\n      \"pmids\": [\"8241958\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1997,\n      \"finding\": \"Detailed structural analysis of T. gondii GPI glycolipid antigens reveals two main structures: (EtNP)Man-Man-(GalNAc)Man-GlcN-PI and (EtNP)Man-Man-(Glc-GalNAc)Man-GlcN-PI. Only GPIs bearing the unique glucose-N-acetylgalactosamine side branch are immunogenic in humans, and this structure is widely distributed among T. gondii isolates.\",\n      \"method\": \"Radiolabeling, chemical/exoglycosidase treatments, HPAEC, gel filtration, lectin affinity chromatography, 2D 1H-NMR, FAB-MS, methylation linkage analysis\",\n      \"journal\": \"Journal of molecular biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — multiple orthogonal structural methods including NMR, MS, and chemical analyses establishing GPI structure and immunogenicity determinant\",\n      \"pmids\": [\"9102470\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1998,\n      \"finding\": \"SAG1 anchored by a transmembrane domain (SAG1-TM) instead of its natural GPI anchor is correctly folded and targeted to the parasite surface, indicating the GPI anchor does not determine surface localization or 3D structure. However, GPI-anchored SAG1 forms high-molecular-weight complexes that SAG1-TM does not, suggesting complex formation is one role of the GPI anchor. SAG1-TM has a significantly lower in vitro half-life than SAG1-GPI, suggesting the GPI anchor stabilizes the protein against degradation.\",\n      \"method\": \"Stable transfection with transmembrane-domain replacement constructs, conformational epitope immunoassay, native gel analysis, in vitro half-life measurement\",\n      \"journal\": \"Journal of cell science\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — genetic replacement experiment in T. gondii with multiple functional readouts, single lab\",\n      \"pmids\": [\"9394009\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1999,\n      \"finding\": \"During host cell invasion, T. gondii selectively excludes transmembrane-anchored host proteins at the moving junction from incorporation into the parasitophorous vacuole (PV), while GPI-anchored proteins (Sca-1, CD55, GPI-anchored ICAM-1) and membrane lipids are readily incorporated. This selective exclusion depends on the membrane anchoring mechanism (transmembrane vs. GPI), not extracellular domain differences, as shown using ICAM-1 variants differing only in anchor type.\",\n      \"method\": \"Host cell invasion assay with recombinant ICAM-1 constructs (transmembrane, GPI-anchored, cytoplasmic tail deletion), fluorescence microscopy, flow cytometry\",\n      \"journal\": \"The Journal of experimental medicine\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — reciprocal anchor-swap experiment with multiple engineered constructs establishing mechanism of selective exclusion at moving junction\",\n      \"pmids\": [\"10601353\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2001,\n      \"finding\": \"The GPI anchor of SAG1 (P30) exists in two glycoforms: Man-Man-[GalNAc]Man-GlcN-PI and Man-Man-[Glc-GalNAc]Man-GlcN-PI. Core modification of the anchor with GalNAc occurs prior to protein transfer, as the modifications match free GPI glycolipid precursors. The sn-2 position of the diacylglycerol lipid contains almost exclusively a C18:1 acyl chain.\",\n      \"method\": \"Protease digestion, phase partitioning, gel filtration, HPEIC, exoglycosidase treatment, MALDI-TOF-MS, FAB-MS, methylation linkage analysis\",\n      \"journal\": \"Molecular and biochemical parasitology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — multiple orthogonal structural methods establishing two GPI glycoforms on SAG1 with precise structural characterization\",\n      \"pmids\": [\"11522346\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2002,\n      \"finding\": \"Clostridium septicum alpha-toxin, which targets GPI-anchored proteins, is active against T. gondii tachyzoites (EC50 = 0.2 nM) and binds to parasite GPI-anchored proteins including SAG3 and SAG1. Toxin treatment disrupts the tight connection between plasma membrane and inner membrane complex, causes massive plasma membrane outward extension, and induces ER swelling (vacuolating effect).\",\n      \"method\": \"Cytotoxicity assay, ultrastructural electron microscopy, toxin binding/co-immunoprecipitation with GPI-anchored proteins\",\n      \"journal\": \"Infection and immunity\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct functional assay with morphological readout and protein binding identification, single lab\",\n      \"pmids\": [\"12117945\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"RH\"}]\n    },\n    {\n      \"year\": 2002,\n      \"finding\": \"The 1.7 Å crystal structure of SAG1 reveals a homodimeric configuration in which the dimeric interface is formed by an extended beta-sheet creating a deep groove lined with positively charged amino acids, proposed to serve as a sulfated proteoglycan-binding site mediating promiscuous host cell attachment.\",\n      \"method\": \"X-ray crystallography at 1.7 Å resolution\",\n      \"journal\": \"Nature structural biology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — high-resolution crystal structure with functional inference from structural features\",\n      \"pmids\": [\"12091874\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2003,\n      \"finding\": \"T. gondii GPIs purified from tachyzoites, as well as their core glycans, induce TNF-alpha production in macrophages via NF-κB activation. A synthetic GPI lacking its lipid moiety (GPIa) retains the same activity, but a synthetic GPI with dialkylglycerol instead of diacylglycerol (GPIb) does not induce TNF-alpha and instead inhibits GPI-induced TNF-alpha production, indicating the lipid moiety regulates signaling.\",\n      \"method\": \"Macrophage stimulation assays, TNF-alpha ELISA, NF-κB reporter assay, synthetic GPI analogues\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — structure-activity relationship established using chemically synthesized GPI variants with multiple functional readouts\",\n      \"pmids\": [\"12815041\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2006,\n      \"finding\": \"Using hypotonically permeabilized T. gondii tachyzoites and PI-PLC topology assays, early GPI intermediates (GlcN-PI and GlcNAc-PI) are predominantly localized in the cytoplasmic leaflet of the ER. More mannosylated and side-chain-modified GPI intermediates are also preferentially localized in the cytoplasmic leaflet. A new early intermediate with inositol acylation was identified, confirming inositol acylation occurs in T. gondii before mannosylation.\",\n      \"method\": \"Hypotonic permeabilization, GDP-[3H]mannose/UDP-[3H]GlcNAc radiolabeling, PI-PLC topology assay, protease protection assays, electron microscopy\",\n      \"journal\": \"Eukaryotic cell\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — direct topology determination using permeabilization and PI-PLC treatment with multiple controls and radiolabeled intermediates\",\n      \"pmids\": [\"16896225\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"T. gondii GPI biosynthesis involves inositol acylation (acyl-CoA-dependent, inhibited by PMSF) before mannosylation, followed by inositol deacylation (inhibited by both PMSF and DFP) of fully mannosylated intermediates. Inositol deacylation is required before addition of the GalNAc side chain to the first mannose. The diacylglycerol at the sn-2 position contains almost exclusively C18:1 acyl chain.\",\n      \"method\": \"In vitro GPI biosynthesis with serine protease inhibitors (PMSF, DFP), radiolabeling, chromatographic analysis of GPI intermediates\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 1 / Strong — in vitro pathway dissection with specific inhibitors identifying novel intermediates and establishing biosynthetic ordering\",\n      \"pmids\": [\"17804418\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"T. gondii GPIs activate TLR4-expressing cells (and both TLR2 and TLR4 when core glycan and lipid moieties are separated). MyD88 is absolutely required for TNF-alpha production by macrophages exposed to T. gondii GPIs, whereas TLR2, TLR4, and CD14 individually are not absolutely required. TNF-alpha response is completely abrogated only in TLR2/4-double-deficient macrophages.\",\n      \"method\": \"Reporter gene assay in CHO cells expressing TLRs, TNF-alpha production in macrophages from knockout mice (TLR2-/-, TLR4-/-, TLR2/4-/-, MyD88-/-, CD14-/-), mouse infection model\",\n      \"journal\": \"Journal of immunology\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — genetic epistasis with multiple knockout mouse strains establishing signaling pathway requirements, replicated across multiple cell types\",\n      \"pmids\": [\"17617606\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"ME-49\"}]\n    },\n    {\n      \"year\": 2007,\n      \"finding\": \"Fatty acids (myristic and palmitic acid) co-extracted with T. gondii GPIs inhibit GPI-induced TNF-alpha production in macrophages by inhibiting tyrosine phosphorylation of cytoplasmic proteins and NF-κB activation through a peroxisome proliferator-activated receptor-independent pathway, following rapid entry into the macrophage cytoplasm.\",\n      \"method\": \"TLC separation, macrophage stimulation, TNF-alpha ELISA, NF-κB activation assay, tyrosine phosphorylation assay, PPAR inhibitor treatment\",\n      \"journal\": \"Infection and immunity\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — mechanistic pathway dissection with multiple assays in a single lab study\",\n      \"pmids\": [\"17387164\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2008,\n      \"finding\": \"TgSUB1 is a GPI-anchored protein targeted to the micronemes rather than the parasite surface. The TgSUB1 prodomain is responsible and sufficient for microneme trafficking (demonstrated by targeting a reporter protein to micronemes). Accurate microneme targeting requires correct folding/context since the prodomain expressed alone cannot reach micronemes. This identifies TgSUB1 as a GPI-anchored protein that bypasses the default GPI-dependent surface trafficking pathway.\",\n      \"method\": \"TgSUB1 deletion constructs, chimeric fusion proteins with reporter, immunofluorescence localization, trafficking assays\",\n      \"journal\": \"Traffic\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — domain deletion and chimera experiments establishing microneme targeting signal, single lab\",\n      \"pmids\": [\"18532988\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2009,\n      \"finding\": \"T. gondii GPIs up-regulate MHC class I and class II molecules on primary murine macrophages (both unstimulated and IFN-gamma-stimulated), in contrast to viable parasite infection which blocks MHC up-regulation. GPI-induced MHC class I up-regulation is abrogated by the NF-κB inhibitor SN50, but GPIs do not activate STAT1. MHC up-regulation enhances antigen presentation to CD8+ lymphocytes.\",\n      \"method\": \"Flow cytometry for surface MHC, NF-κB inhibitor (SN50) treatment, STAT1 activation assay, CD8+ T cell antigen presentation assay\",\n      \"journal\": \"Innate immunity\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — pathway dissection with specific inhibitor and multiple readouts, single lab\",\n      \"pmids\": [\"19201822\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"T. gondii GPIs fail to block apoptosis triggered via extrinsic or intrinsic pathways in human-derived cells, as assessed by caspase-3/7 activity, phosphatidylserine exposure, and DNA strand breaks. This is a negative result: T. gondii GPIs are not involved in host cell survival or apoptosis induction.\",\n      \"method\": \"Caspase-3/7 activity assay, annexin V (phosphatidylserine) staining, TUNEL (DNA strand break assay), extrinsic/intrinsic apoptosis triggers\",\n      \"journal\": \"Apoptosis\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — multiple orthogonal apoptosis assays establishing negative result, single lab\",\n      \"pmids\": [\"17252196\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2010,\n      \"finding\": \"TgSUB1 lacking its GPI anchor (Δsub1::ΔGPISUB1) only partially restores microneme protein (MIC2, MIC4, M2AP) surface processing and cell attachment/gliding compared to full-length TgSUB1 complementation, establishing that the GPI anchor of TgSUB1 contributes to its full proteolytic function at the parasite surface.\",\n      \"method\": \"Genetic complementation (full-length vs. GPI-deleted TgSUB1), microneme protein processing assays, cell attachment/gliding efficiency assays, mouse virulence\",\n      \"journal\": \"Cellular microbiology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — complementation experiment with GPI-deletion variant establishing functional contribution of GPI anchor to protease activity, single lab\",\n      \"pmids\": [\"20678172\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2014,\n      \"finding\": \"Virulent (RH) and avirulent (PTG) T. gondii strains differ in GPI biosynthesis: PTG has ~34% of the protein-free GPIs and ~70% of the GPI-anchored proteins of RH, with lower rates of protein N-glycosylation, and two of six GPI intermediates are significantly reduced in PTG. Despite these quantitative differences, individual mature GPI species from both strains induce comparable TNF-alpha and IL-12p40 secretion via TLR4/MyD88-dependent NF-κBp65 activation.\",\n      \"method\": \"Metabolic radioactive labeling, GPI quantification, ELISA (TNF-alpha, IL-12p40), NF-κBp65 activation assay, TLR4/MyD88 pathway analysis\",\n      \"journal\": \"PloS one\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — comparative strain analysis with quantitative GPI profiling and functional assays, single lab\",\n      \"pmids\": [\"24489660\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"RH\"}, {\"species\": \"Toxoplasma gondii\", \"strain\": \"PTG\"}]\n    },\n    {\n      \"year\": 2016,\n      \"finding\": \"Depletion of host cell GPI (using GPI-deficient host cells) alters evacuole formation during T. gondii invasion, causing excess evacuole formation, independently of the evacuole-inhibiting effect of cholesterol depletion. Excess evacuoles led to enhanced recruitment of host mitochondria and endoplasmic reticulum to parasitophorous vacuoles and stimulated parasite growth.\",\n      \"method\": \"GPI-deficient host cells, cholesterol depletion (methyl-beta-cyclodextrin), fluorescence microscopy of evacuoles, parasite growth assay\",\n      \"journal\": \"Parasitology international\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — genetic loss-of-function (GPI-deficient host cells) with functional readouts, single lab\",\n      \"pmids\": [\"27217289\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2018,\n      \"finding\": \"Using CRISPR/Cas9 deletion of 17 glycogenes combined with MS-based glycomics, the glycosyltransferases for assembling novel nuclear O-Fuc-type and cell surface Glc-Fuc-type O-glycans were identified. GPI-anchor assembly was among the pathways mapped, with several glycogenes shown to be important for in vitro growth.\",\n      \"method\": \"Double-CRISPR/Cas9 gene editing, MS-based glycomics, in vitro growth assays\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — systematic CRISPR screen with glycomics validation, single lab; GPI pathway was part of a broader screen\",\n      \"pmids\": [\"30463938\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2019,\n      \"finding\": \"Immunoelectron microscopy using quick-freezing and freeze-fracture labeling (QF-FRL) of living T. gondii shows SAG1 (GPI-anchored) labeling exclusively in the exoplasmic leaflet of the plasma membrane with no labeling in the inner membrane complex. Point pattern analysis demonstrates SAG1 is randomly distributed and does not form significant microdomains (lipid rafts) in the native T. gondii plasma membrane.\",\n      \"method\": \"Quick-freezing/freeze-fracture labeling immunoelectron microscopy, point pattern statistical analysis\",\n      \"journal\": \"Histochemistry and cell biology\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — direct nanoscale localization method with quantitative spatial analysis, single lab\",\n      \"pmids\": [\"31542792\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2022,\n      \"finding\": \"Depletion of the apicoplast SUF pathway (TgNFS2 or TgSUFC) impairs proper targeting of GPI-anchored proteins and reduces parasite motility, demonstrating that isoprenoid precursors generated by the apicoplast (via iron-sulfur cluster-dependent enzymes) are necessary for GPI-anchor biosynthesis and GPI-anchored protein surface targeting.\",\n      \"method\": \"Conditional knockout (TgNFS2), direct knockout (TgSUFC), immunofluorescence of GPI-anchored protein localization, motility assays, fatty acid supplementation rescue experiments\",\n      \"journal\": \"The Journal of biological chemistry\",\n      \"confidence\": \"Medium\",\n      \"confidence_rationale\": \"Tier 2 / Moderate — conditional genetic knockouts with specific functional readouts, single lab, multiple orthogonal assays\",\n      \"pmids\": [\"35810787\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2024,\n      \"finding\": \"PIGJ (a glycosyltransferase) adds the first GalNAc to the GPI sidechain, and PIGE subsequently adds a terminal glucose. Deletion of PIGJ eliminates the entire GPI sidechain and increases virulence during primary and secondary infections. PIGJ-deletion mutants show enhanced binding by scavenger receptor CD36 and galectin-3. Protection of galectin-3 knockout mice from lethality caused by ΔPIGJ parasites indicates that the virulence phenotype of sidechain-deficient parasites is sidechain- and galectin-3-dependent.\",\n      \"method\": \"CRISPR/Cas9 gene deletion, serology, mass spectrometry glycan analysis, mouse infection models (primary and secondary), CD36 binding assay, galectin-3 KO mouse experiments, immunofluorescence\",\n      \"journal\": \"mBio\",\n      \"confidence\": \"High\",\n      \"confidence_rationale\": \"Tier 2 / Strong — genetic knockouts of both glycosyltransferases with MS structural validation, multiple in vivo models, and mechanistic pathway dissection via galectin-3 KO\",\n      \"pmids\": [\"39302131\"],\n      \"is_preprint\": false,\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    }\n  ],\n  \"current_model\": \"Toxoplasma gondii GPI anchors are complex glycolipids biosynthesized in the ER via sequential glycosylation of phosphatidylinositol, with inositol acylation preceding mannosylation and subsequent deacylation required before GalNAc side-chain addition; mature GPIs are predominantly assembled in the cytoplasmic leaflet of the ER and depend on apicoplast-generated isoprenoid precursors for proper targeting of GPI-anchored surface proteins. The GPI sidechain (assembled by glycosyltransferases PIGJ and PIGE) modulates host-parasite interactions — its absence increases virulence and enhances scavenger receptor CD36 and galectin-3 binding — while free GPIs and GPI-anchored proteins activate innate immune responses via TLR2/TLR4 and MyD88-dependent NF-κB signaling to induce TNF-alpha and modulate MHC expression, and the GPI anchor itself stabilizes surface proteins, promotes high-molecular-weight complex formation, and determines selective exclusion of parasite-surface proteins at the moving junction during host cell invasion.\"\n}\n```","stage2_raw":"{\n  \"mechanistic_narrative\": \"In Toxoplasma gondii, GPI (glycosylphosphatidylinositol) anchors are complex glycolipids assembled in the ER that attach abundant surface antigens such as SAG1/P30, gp23, and SAG3 to the parasite plasma membrane and simultaneously function as potent immunomodulators of the host [#1, #2, #9]. Biosynthesis proceeds by sequential glycosylation of phosphatidylinositol using GDP-mannose, UDP-GlcNAc, and Dol-P-mannose, producing a conserved core (EtNP-Man-Man-Man-GlcN-PI) that in T. gondii is uniquely elaborated with a GalNAc side branch and, in one immunogenic glycoform, a terminal glucose (Glc-GalNAc) [#0, #3, #6]. The pathway is topologically and enzymatically ordered: inositol acylation precedes mannosylation, and inositol deacylation of fully mannosylated intermediates is required before GalNAc side-chain addition, with early intermediates residing predominantly in the cytoplasmic leaflet of the ER [#10, #11]; the terminal side-chain sugars are installed by the glycosyltransferases PIGJ (first GalNAc) and PIGE (terminal glucose) [#23]. The anchor is dispensable for surface localization and folding but stabilizes surface proteins against degradation and promotes high-molecular-weight complex formation [#4], and the GPI-versus-transmembrane anchoring mode governs selective sorting of proteins at the moving junction during invasion [#5]. Purified T. gondii GPIs and their core glycans drive innate immune activation, inducing TNF-alpha and IL-12 and up-regulating MHC via TLR2/TLR4- and MyD88-dependent NF-\\u03baB signaling, with the diacylglycerol lipid moiety required for signaling [#9, #12, #15]. The GalNAc-containing side chain assembled by PIGJ restrains virulence, as its deletion increases virulence and enhances CD36 and galectin-3 binding in a galectin-3-dependent manner [#23].\",\n  \"teleology\": [\n    {\n      \"year\": 1992,\n      \"claim\": \"Establishing that T. gondii assembles its own GPI anchors defined the biosynthetic logic: a cell-free system showed the core glycan is built by sequential glycosylation of phosphatidylinositol from sugar-nucleotide donors, and free glycolipid precursors sharing the P30 anchor backbone identified the pathway's intermediates.\",\n      \"evidence\": \"cell-free biosynthesis with radiolabeled sugar nucleotides plus enzymatic/chemical analysis of free glycolipids\",\n      \"pmids\": [\"1328232\", \"1534801\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"individual glycosyltransferases not yet identified\", \"subcellular topology of assembly not defined\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1993,\n      \"claim\": \"Direct structural analysis of surface glycoproteins confirmed GPI as the physiological membrane anchor of parasite antigens, showing gp23 carries the conserved EtNP-Man3-GlcN-PI core modified by beta-GalNAc and is releasable by PI-PLC.\",\n      \"evidence\": \"metabolic radiolabeling, PI-PLC/PLA2 and glycosidase digestion, immunoprecipitation of gp23\",\n      \"pmids\": [\"8241958\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"functional role of the GalNAc modification unknown\", \"timing of side-chain addition relative to protein transfer unresolved\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1997,\n      \"claim\": \"Complete structural characterization resolved the two major GPI glycoforms and pinpointed the Glc-GalNAc side branch as the determinant of human immunogenicity, linking a specific glycan feature to host recognition.\",\n      \"evidence\": \"NMR, FAB-MS, methylation linkage analysis, lectin chromatography of purified GPIs\",\n      \"pmids\": [\"9102470\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"enzymes adding Glc and GalNAc not identified\", \"receptor(s) recognizing the immunogenic branch unknown\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1998,\n      \"claim\": \"Replacing the SAG1 GPI anchor with a transmembrane domain dissected the anchor's function from targeting, showing GPI is not required for surface delivery or folding but stabilizes the protein and enables high-molecular-weight complex formation.\",\n      \"evidence\": \"transmembrane-domain swap transfection with conformational assay, native gels, half-life measurement\",\n      \"pmids\": [\"9394009\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"molecular basis of complex formation unknown\", \"single lab, single reporter protein\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 1999,\n      \"claim\": \"Anchor-swap experiments during invasion established that the membrane-anchoring mode itself, not extracellular domain identity, dictates selective inclusion/exclusion of proteins at the moving junction, with GPI-anchored proteins incorporated into the parasitophorous vacuole.\",\n      \"evidence\": \"invasion assays with ICAM-1 variants differing only in anchor type, microscopy, flow cytometry\",\n      \"pmids\": [\"10601353\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"molecular machinery sensing anchor type at the junction not identified\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2001,\n      \"claim\": \"Structural mapping of the SAG1 anchor showed core GalNAc modification occurs before protein transfer (matching free precursors) and defined the C18:1-dominant sn-2 lipid, ordering side-chain addition within the pathway.\",\n      \"evidence\": \"MALDI-TOF/FAB-MS, exoglycosidase treatment, methylation analysis of purified SAG1 anchor\",\n      \"pmids\": [\"11522346\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"deacylation requirement for side-chain addition not yet established\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2002,\n      \"claim\": \"A pharmacological probe (alpha-toxin) and the SAG1 crystal structure connected GPI-anchored proteins to plasma-membrane/IMC integrity and to host attachment via a positively charged proteoglycan-binding groove.\",\n      \"evidence\": \"alpha-toxin cytotoxicity/EM/binding assays and 1.7 \\u00c5 X-ray crystallography of SAG1\",\n      \"pmids\": [\"12117945\", \"12091874\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"direct proteoglycan binding by SAG1 not demonstrated\", \"toxin-binding proteins beyond SAG1/SAG3 not fully catalogued\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"RH\"}]\n    },\n    {\n      \"year\": 2003,\n      \"claim\": \"Structure-activity analysis with synthetic GPI analogues identified the diacylglycerol lipid moiety as required for TNF-alpha induction via NF-\\u03baB, establishing GPI as a defined immunostimulatory ligand.\",\n      \"evidence\": \"macrophage stimulation, TNF-alpha ELISA, NF-\\u03baB reporter with synthetic GPI variants (GPIa/GPIb)\",\n      \"pmids\": [\"12815041\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"surface receptor not identified in this study\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2006,\n      \"claim\": \"Topology assays placed early and modified GPI intermediates predominantly in the cytoplasmic ER leaflet and confirmed inositol acylation precedes mannosylation, defining where and in what order assembly occurs.\",\n      \"evidence\": \"permeabilized tachyzoites, radiolabeled intermediates, PI-PLC topology and protease-protection assays\",\n      \"pmids\": [\"16896225\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"flippase/transfer to exoplasmic leaflet not defined\", \"enzymes catalyzing each step not identified\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"Inhibitor-based pathway dissection established a two-step acylation/deacylation logic in which inositol deacylation of fully mannosylated intermediates is required before GalNAc side-chain addition, resolving the biosynthetic ordering.\",\n      \"evidence\": \"in vitro GPI biosynthesis with serine protease inhibitors (PMSF, DFP) and chromatographic intermediate analysis\",\n      \"pmids\": [\"17804418\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"identity of the acyltransferase and deacylase not established\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2007,\n      \"claim\": \"Genetic epistasis with knockout mice mapped GPI-driven TNF-alpha induction to MyD88 as absolutely required, with TLR2 and TLR4 acting redundantly, and identified co-extracted fatty acids as endogenous negative regulators of the response.\",\n      \"evidence\": \"TLR reporter assays and TNF-alpha production in TLR2/4/MyD88/CD14 knockout macrophages; fatty-acid inhibition assays\",\n      \"pmids\": [\"17617606\", \"17387164\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"direct GPI-TLR binding not shown\", \"in vivo relevance of fatty-acid regulation not established\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"ME-49\"}]\n    },\n    {\n      \"year\": 2009,\n      \"claim\": \"GPIs were shown to up-regulate MHC class I and II via NF-\\u03baB (SN50-sensitive, STAT1-independent), extending GPI immunomodulation to enhanced antigen presentation to CD8+ T cells and contrasting with immune evasion by live parasites.\",\n      \"evidence\": \"flow cytometry of surface MHC with SN50 inhibitor, STAT1 assay, CD8+ presentation assay\",\n      \"pmids\": [\"19201822\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"single lab\", \"receptor upstream of NF-\\u03baB not defined in this context\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2010,\n      \"claim\": \"Complementary functional studies bounded GPI activity: GPIs do not modulate host apoptosis (a negative result), while the TgSUB1 GPI anchor was shown to contribute to full surface proteolytic processing of microneme proteins.\",\n      \"evidence\": \"caspase/annexin/TUNEL apoptosis panel; genetic complementation of TgSUB1 with GPI-deleted variant and processing/gliding assays\",\n      \"pmids\": [\"17252196\", \"20678172\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"mechanism by which the anchor supports TgSUB1 activity unresolved\", \"single lab studies\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2008,\n      \"claim\": \"Characterization of TgSUB1 revealed that some GPI-anchored proteins bypass default surface trafficking, with the prodomain sufficient to redirect a GPI-anchored protein to micronemes.\",\n      \"evidence\": \"deletion/chimera constructs and immunofluorescence trafficking assays\",\n      \"pmids\": [\"18532988\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"sorting machinery recognizing the prodomain not identified\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2014,\n      \"claim\": \"Comparative profiling of virulent RH and avirulent PTG strains showed quantitative differences in GPI and GPI-protein abundance yet comparable per-species cytokine induction, indicating GPI immunostimulatory potency is intrinsic to the mature structure rather than to strain-specific quantity.\",\n      \"evidence\": \"metabolic labeling, GPI quantification, TNF-alpha/IL-12 ELISA, TLR4/MyD88/NF-\\u03baB analysis in RH vs PTG\",\n      \"pmids\": [\"24489660\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"genetic basis of biosynthetic quantitative differences not defined\", \"in vivo consequence of GPI abundance differences not tested\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"RH\"}, {\"species\": \"Toxoplasma gondii\", \"strain\": \"PTG\"}]\n    },\n    {\n      \"year\": 2016,\n      \"claim\": \"Loss-of-function in host GPI showed host-cell GPI restrains evacuole formation during invasion, and its absence enhances organelle recruitment to the vacuole and parasite growth, revealing a host-side contribution of GPI to the invasion interface.\",\n      \"evidence\": \"GPI-deficient host cells with cholesterol-depletion controls, evacuole microscopy, growth assays\",\n      \"pmids\": [\"27217289\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"molecular link between host GPI and evacuole control unknown\", \"distinct from parasite GPI biology\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2018,\n      \"claim\": \"A systematic CRISPR/glycomics screen placed GPI-anchor assembly within the broader parasite glycosylation network and flagged glycogenes important for in vitro growth, beginning the transition from biochemical intermediates to defined genes.\",\n      \"evidence\": \"double-CRISPR/Cas9 deletion of 17 glycogenes with MS glycomics and growth assays\",\n      \"pmids\": [\"30463938\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"specific GPI-pathway gene functions not individually resolved here\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2019,\n      \"claim\": \"Nanoscale localization confirmed the GPI-anchored SAG1 resides exclusively in the exoplasmic leaflet of the plasma membrane and is randomly distributed, arguing against lipid-raft microdomain organization for this anchor.\",\n      \"evidence\": \"quick-freeze/freeze-fracture immuno-EM with point-pattern statistical analysis\",\n      \"pmids\": [\"31542792\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"whether other GPI proteins cluster not tested\", \"single lab\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2022,\n      \"claim\": \"Conditional knockouts of apicoplast SUF-pathway components linked apicoplast-derived isoprenoid precursors to proper GPI-anchor biosynthesis and surface targeting of GPI-anchored proteins, connecting organellar metabolism to the GPI pathway.\",\n      \"evidence\": \"conditional (TgNFS2) and direct (TgSUFC) knockouts, GPI-protein localization, motility, fatty-acid rescue\",\n      \"pmids\": [\"35810787\"],\n      \"confidence\": \"Medium\",\n      \"gaps\": [\"precise isoprenoid-dependent step in GPI assembly not pinpointed\", \"single lab\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": 2024,\n      \"claim\": \"Gene-level assignment of the side-chain glycosyltransferases PIGJ and PIGE, combined with in vivo infection and receptor-binding assays, established that the GalNAc-initiated side chain restrains virulence in a galectin-3-dependent manner and shapes CD36/galectin-3 recognition.\",\n      \"evidence\": \"CRISPR deletion of PIGJ/PIGE, MS glycan analysis, primary/secondary mouse infection, CD36 binding, galectin-3 KO mice\",\n      \"pmids\": [\"39302131\"],\n      \"confidence\": \"High\",\n      \"gaps\": [\"upstream regulation of side-chain synthesis unknown\", \"human relevance of the CD36/galectin-3 axis not tested\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    },\n    {\n      \"year\": null,\n      \"claim\": \"The complete enzymatic roster for T. gondii GPI core and lipid remodeling (mannosyltransferases, the acyltransferase/deacylase, EtNP transferase, and the GPI transamidase) and the receptor(s) directly engaging GPI to trigger TLR/MyD88 signaling remain unassigned in the available corpus.\",\n      \"evidence\": \"\",\n      \"pmids\": [],\n      \"confidence\": \"Low\",\n      \"gaps\": [\"most core biosynthetic genes not individually identified\", \"direct GPI-receptor binding not demonstrated\", \"structural model of the GPI transamidase complex absent\"],\n      \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]\n    }\n  ],\n  \"mechanism_profile\": {\n    \"molecular_activity\": [\n      {\"term_id\": \"GO:0000030\", \"supporting_discovery_ids\": [0, 3, 6, 11, 23], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0006506\", \"supporting_discovery_ids\": [0, 1, 2, 10, 11], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0008194\", \"supporting_discovery_ids\": [23], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]}\n    ],\n    \"localization\": [\n      {\"term_id\": \"GO:0005789\", \"supporting_discovery_ids\": [10, 11], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0009986\", \"supporting_discovery_ids\": [2, 4, 21], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0031362\", \"supporting_discovery_ids\": [2, 4, 21], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0020009\", \"supporting_discovery_ids\": [14], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]}\n    ],\n    \"pathway\": [\n      {\"term_id\": \"GO:0006506\", \"supporting_discovery_ids\": [0, 1, 3, 6, 10, 11, 23], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0007249\", \"supporting_discovery_ids\": [9, 12, 15, 18], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]},\n      {\"term_id\": \"GO:0002224\", \"supporting_discovery_ids\": [12], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"ME-49\"}]},\n      {\"term_id\": \"GO:0030260\", \"supporting_discovery_ids\": [5, 19], \"attribution\": [{\"species\": \"Toxoplasma gondii\", \"strain\": \"\"}]}\n    ],\n    \"complexes\": [\n      \"GPI-anchored SAG1 high-molecular-weight surface complexes [#4]\"\n    ],\n    \"partners\": [\n      \"SAG1/P30 (GPI-anchored surface antigen) [#1, #4, #6, #8, #21]\",\n      \"gp23 (GPI-anchored surface glycoprotein) [#2]\",\n      \"SAG3 (GPI-anchored, alpha-toxin binding) [#7]\",\n      \"TgSUB1 (GPI-anchored subtilisin, microneme-targeted) [#14, #17]\",\n      \"TLR2 / TLR4 (innate immune receptors of GPI) [#12]\",\n      \"MyD88 (required adaptor for GPI-induced TNF-alpha) [#12]\",\n      \"PIGJ (GalNAc side-chain glycosyltransferase) [#23]\",\n      \"PIGE (terminal glucose glycosyltransferase) [#23]\",\n      \"CD36 (scavenger receptor binding sidechain-deficient parasites) [#23]\",\n      \"Galectin-3 (host lectin mediating sidechain-dependent virulence) [#23]\",\n      \"TgNFS2 / TgSUFC (apicoplast SUF pathway supporting GPI-protein targeting) [#22]\"\n    ],\n    \"other_free_text\": [\n      \"T. gondii GPI core: EtNP-Man-Man-Man-GlcN-PI, uniquely modified with a beta-GalNAc side branch and, in one glycoform, a terminal Glc-GalNAc; sn-2 lipid is almost exclusively C18:1 [#3, #6, #11].\",\n      \"Biosynthetic ordering: inositol acylation precedes mannosylation; inositol deacylation of fully mannosylated intermediates precedes GalNAc side-chain addition; early intermediates reside in the cytoplasmic ER leaflet [#10, #11].\",\n      \"The Glc-GalNAc side branch is the human immunogenicity determinant [#3]; the diacylglycerol lipid moiety is required for TNF-alpha induction [#9].\",\n      \"GPI anchor stabilizes surface proteins and drives high-MW complex formation but is not required for folding/surface targeting [#4]; anchor type dictates protein sorting at the moving junction [#5].\",\n      \"Host-cell GPI restrains evacuole formation during invasion; its depletion enhances organelle recruitment and parasite growth [#19].\",\n      \"T. gondii GPIs do not modulate host apoptosis [#16].\"\n    ]\n  },\n  \"strain_conflicts\": []\n}","audit_flag":null,"strains":[{"strain":"ME49","gene_id":"TGME49_283780","name":"GPI","product":"glucose-6-phosphate isomerase GPI","uniprot_acc":"S8FBB6","chromosome":"TGME49_chrV","start_pos":2973954},{"strain":"GT1","gene_id":"TGGT1_283780","name":null,"product":"glucose-6-phosphate isomerase GPI","uniprot_acc":null,"chromosome":"TGGT1_chrV","start_pos":2809762},{"strain":"RH88","gene_id":"TGRH88_027150","name":null,"product":"glucose-6-phosphate isomerase GPI","uniprot_acc":null,"chromosome":"CM023086","start_pos":3012305},{"strain":"VEG","gene_id":"TGVEG_283780","name":null,"product":"glucose-6-phosphate isomerase GPI","uniprot_acc":"V4Z938","chromosome":"TGVEG_chrV","start_pos":2930189}]}