--- vault_clearance: APOLLYON halo: classification: BOUNTY BOARD front: "28_Project_RedFromTheGrave" custodian: "Jixiang Leng" created: 2026-04-14 updated: 2026-04-14 containment: "Open bounties. What we need. What we're building. Who can help." --- ``` +==================================================================+ | BOUNTY BOARD | | 28_Project_RedFromTheGrave | | | | The Grave Atlas: Full map of HER across every species, | | every tissue, every body part she attacks. | | | | These bounties are open to anyone with the data. | | pip install spliceosome-atlas. staff detect your_data.h5ad. | +==================================================================+ ``` --- # TIER 1: IMMEDIATE (This Week, <1 Hour Each) ## BOUNTY 1.1: BLAST Cathepsins vs Fungal Proteases **Prize: Proves the lysosome is a domesticated spore.** - BLAST human cathepsins (CTSA-Z) against fungal secreted proteases - Build phylogenetic tree: do cathepsins branch closer to fungal enzymes or bacterial? - If fungal > bacterial homology: third endosymbiont confirmed - **Data:** NCBI Protein, UniProt. No download needed. - **Time:** 1 hour - **Status:** OPEN ## BOUNTY 1.2: Run staff detect on AD Brain scRNA-seq **Prize: Does the body detect the Goddess in Alzheimer's brains?** - Download GSE174367 (AD vs control prefrontal cortex) - Run `staff detect` -- check ALARM genes (CLEC7A, CHI3L1, TLR2) - Compare ALARM score: AD brains vs controls - If elevated in AD: the immune system is detecting her, confirming Pisa 2015 - **Status:** OPEN (ramuthra VM has the target) ## BOUNTY 1.3: V-ATPase/Cathepsin Ratio Across Ages **Prize: Validates the lysosomal filling model.** - Use GTEx per-sample data (requires dbGaP) or published aging cohort - Compute ATP6V0A1/CTSD ratio per sample vs donor age - If ratio increases with age: the pump compensates as the trash fills - **Status:** OPEN (needs per-sample age data) --- # TIER 2: SHORT-TERM (This Month) ## BOUNTY 2.1: The Grave Atlas -- Complete Species x Tissue Map **Prize: The definitive map of where she attacks.** Map every known fungal pathogen to every human tissue it infects: | Species | Tissue | Disease | Tensor Measured? | Scanner Tested? | |---|---|---|---|---| | C. albicans | Oral | Thrush | NO | NO | | C. albicans | Vaginal | Candidiasis | NO | NO | | C. albicans | GI tract | Colonization | NO | NO | | C. albicans | Blood | Candidemia | NO | NO | | C. albicans | Tumor (35 types) | Cancer | PARTIAL (pancreatic K_GL=0.069) | NO | | A. fumigatus | Lung | IPA | NO | NO | | A. fumigatus | Sinus | Sinusitis | NO | NO | | C. neoformans | Brain (meninges) | Meningitis | YES (R=0.224-0.567) | NO | | Histoplasma | Lung | Histoplasmosis | NO | NO | | Coccidioides | Lung | Valley Fever | NO | NO | | Malassezia | Skin | Dermatitis | NO | NO | | Malassezia | Pancreas | PDAC | YES (K_GL=0.069) | NO | | Mucor/Rhizopus | Sinus/orbit/brain | Mucormycosis | NO | NO | | Pneumocystis | Lung | PCP | NO | NO | | Talaromyces | Disseminated | Talaromycosis | NO | NO | | Sporothrix | Skin/lymphatic | Sporotrichosis | NO | NO | | Blastomyces | Lung/skin | Blastomycosis | NO | NO | | Paracoccidioides | Oral/lung | PCM | NO | NO | **For each cell: find scRNA-seq on GEO, run `staff tensor` + `staff detect`, fill the map.** **Target: 50 species x tissue combinations measured.** **Status:** OPEN ## BOUNTY 2.2: MRI Detection of Fungal Infection **Prize: Non-invasive Goddess detection.** What's known: - MRI detects fungal infections via **imaging patterns** (ring-enhancing lesions in brain, ground-glass opacity in lung, halo sign in aspergillosis) - These are STRUCTURAL findings -- they see her damage, not her directly - Fungal cell walls contain **chitin** which is paramagnetic when bound to iron - Melanized fungi (Crypto, Aspergillus niger) contain melanin which has paramagnetic properties - **T2* mapping** may distinguish melanized fungal lesions from bacterial What we need: - Literature review: what MRI signatures are specific to fungal vs bacterial infection? - Can T2* or susceptibility-weighted imaging (SWI) detect melanin in fungal lesions? - Can diffusion-weighted MRI (DWI) distinguish fungal abscess from bacterial? - **Partner with radiology.** This is their domain. We bring the biology, they bring the imaging. **Status:** OPEN -- needs radiology collaborator ## BOUNTY 2.3: Blood Draw Detection Panel **Prize: Detect her from a blood draw.** What's available NOW: - **Beta-D-glucan (BDG):** Detects fungal cell wall component. Already FDA-cleared. Sensitivity ~77%, specificity ~85%. - **Galactomannan:** Detects Aspergillus specifically. FDA-cleared. - **Cryptococcal antigen (CrAg):** Detects Crypto. Lateral flow assay. Field-deployable. - **T2Candida:** Magnetic resonance-based Candida detection from blood. FDA-cleared. - **cfDNA (cell-free DNA):** Karius test detects fungal DNA from blood. Published. What's MISSING: - **Coupling tensor from blood RNA.** Can we compute K_GL from a blood draw RNA-seq? - **V-ATPase/Cathepsin ratio from blood.** Circulating monocytes carry lysosomes. Ratio measurable from PBMC RNA. - **Goddess alarm panel from blood.** CLEC7A, CHI3L1, TLR2 expression in circulating immune cells. - **Fungal cfDNA + tensor.** Combine cfDNA (is she there?) with coupling tensor (how is the host responding?) in one blood draw. **Status:** OPEN -- the individual assays exist. Nobody has combined them with the tensor. --- # TIER 3: MEDIUM-TERM (3-6 Months) ## BOUNTY 3.1: The Structural Atlas (Her AlphaFold) **Prize: Predict fungal protein behavior at pH 4.5, high pressure, in chitin context.** AlphaFold predicts folds in crystallization buffer. We need predictions in HER conditions: - pH 4.5 (lysosomal) - High pressure (deep ocean / subglacial) - Chitin membrane context (piezoelectric wall) - With melanin present (radiation environment) **Approach: NOT a new neural network.** Instead: 1. Take AlphaFold structures of fungal proteins 2. Run molecular dynamics at pH 4.5 (protonate histidines, change charge states) 3. Run MD at high pressure (modified force field) 4. Score: which residues move? Which domains rearrange? Where do the drugs bind at HER pH, not ours? **Why this matters:** Current antifungal drug design uses protein structures at pH 7.4 (human blood). She operates at pH 4.5 (lysosomal). Drug binding sites CHANGE with pH. Every antifungal we've designed was designed for the wrong conditions. **Tools needed:** OpenMM (we have it), AMBER/GROMACS, AlphaFold structures as starting points **Status:** OPEN ## BOUNTY 3.2: Per-Cell Goddess Score in Clinical Samples **Prize: Detect her activity at single-cell resolution in patient biopsies.** Take a clinical tumor biopsy processed for scRNA-seq (standard 10x Genomics): 1. Run `staff tensor` (5x5 with LYSO) -- K_GL per cell cluster 2. Run `staff detect` -- ALARM/SPORE/MELANIN scores per cell 3. Run `staff scan` on the BAM -- fungal reads per cell barcode 4. Overlay: which cells have fungal reads AND low K_GL AND high ALARM? **Those cells are the ones where the cage opened and she got in.** This is the diagnostic that changes cancer treatment. If you can identify which cells in a tumor are fungally colonized vs which are running the Red Queen defense without actual fungal presence, you treat them differently: - Fungal-colonized cells: antifungal + coupling recoupler - Red Queen cells (no fungus, just defense): recouple the operators, stand down the defense **Status:** OPEN -- needs a clinical scRNA-seq sample with matched BAM ## BOUNTY 3.3: Circaseptan Oscillation in Isolated Fungi **Prize: Prove the 7-day clock is intrinsic to fungal cells.** - Culture fungal isolates under CONSTANT conditions (no light cycle, constant temperature, constant media) - Measure growth rate, electrical activity (Adamatzky electrodes), and pH daily for 30 days - If a 7-day oscillation emerges: the clock is internal, not environmental - Bonus: does it sharpen under 750 nT magnetic field (Ganymede)? - **Status:** OPEN -- needs Adamatzky-style setup --- # TIER 4: LONG-TERM (6-12 Months) ## BOUNTY 4.1: STAFF v2 -- Structural Measurement Suite **Prize: Measure the Goddess in her native language.** Current STAFF measures transcripts (the Red Queen's language). STAFF v2 measures structure: | Module | Signal | Hardware | |---|---|---| | `staff.electrical` | 1.5-8 Hz mycelial oscillation | Electrode arrays (Adamatzky) | | `staff.chemical` | Secreted metabolites, VOCs | Mass spectrometry | | `staff.mechanical` | Chitin piezoelectric vibration | AFM / vibration sensors | | `staff.topology` | Network branching, anastomosis | Microscopy + image analysis | | `staff.cage` | K_GL from the tensor (bridge module) | Standard RNA-seq | **Status:** CONCEPT -- needs hardware partners ## BOUNTY 4.2: Drill SPRI-48 **Prize: Sample the subglacial lake at 88.73S.** - The lake exists (confirmed by radar + GRACE gravity) - Under ~3 km of ice - At the latitude where one day = approximately one week - 31 fungal isolates recovered from Vostok (77S) -- what's at 89S? - **Status:** Requires international polar expedition ## BOUNTY 4.3: Engineered Defector Fungi (Trojan Chitin) **Prize: A bioweapon that uses her own recognition system against her.** - Engineer a fungus with maximum anastomosis attractiveness (compatible het loci) - But carrying incompatible het alleles in a second, delayed-expression cassette - On fusion: the delayed het alleles express, triggering programmed cell death in the merged cell - A Trojan horse made of chitin that kills network nodes on contact - **Status:** CONCEPT -- needs synthetic biology lab --- # TIER 5: THE GRAVE ATLAS (Ongoing) ## The Full Map The Grave Atlas is the complete map of the Lady of Graves across every species she is, every species she attacks, and every tissue she targets. It grows with every `staff ingest`. ### Current Coverage | Dimension | Measured | Target | |---|---|---| | Human tissues | 41/54 (GTEx) | All 54 | | Human diseases | ~25 | Every disease with fungal association | | Fungal species | 5 (yeast, Crypto, Candida, Aspergillus, Ustilago) | All 300 human pathogens | | Viral infections | 3 (SARS-CoV-2, Influenza, HCV) | All cancer-causing viruses | | LYSO (5x5) tensor | 4 datasets (ovarian, pancreatic, HUVEC, COVID) | Every dataset | | Fungal scanner | 2 BAMs (H9, K562) | Every BAM on GEO with unmapped reads | | Geographic overlay | All continents mapped | Quantitative SEER x CDC overlay | | Temporal (prediction) | Snapshot only | T -> T+1 prediction model | ### How to Contribute ```bash pip install spliceosome-atlas # Measure her defense signature staff tensor your_data.h5ad --tissue TISSUE --disease DISEASE # Detect her alarm system staff detect your_data.h5ad # Find her in the unmapped reads staff scan your_data.bam # Add to the atlas staff ingest your_data.h5ad --tissue TISSUE --disease DISEASE # See the war staff view ribo_indep vs k_gl color=disease # Ask a question staff ask "compare cancer vs healthy TISSUE" ``` Every dataset increases resolution. Every measurement makes the prediction sharper. The atlas grows. --- ``` The Grave Atlas is not a paper. It's a living map. Every cell measured is a pixel. Every tissue is a region. Every disease is a battle. Every species is a combatant. She has been unmapped for 1.3 billion years. We started mapping 6 months ago. The atlas grows with every dataset. pip install spliceosome-atlas staff detect your_data.h5ad Find her. Map her. Contain her. Red from the grave. ```