Further findings about plant derived extracellular vesicles (AI assisted research / PathMap.org)

Hello again community! I have been busy trying to manage life yada yadda but at the same time doing plenty of reading and I am looking for suggestions for how to take the research path I am on and turn it into a useful data graph / dataset.

Currently, PathMap.org hosts a small library of research questions and ideas as I search for ways to accelerate research for cures (I am not a medical professional, and I understand code / variable big data in my own way).

While the ideas are novel and in my opinion, may contain life saving ideas, the platform is limited in its reach and impact.

Last night I watched Homestead Rescue and the guy was talking about how he lost his wife suddenly and quickly to ovarian cancer. I looked at Laura and told her it is sad because what if my ideas about 6-shogaol delivery via Ginger-EVs for Ovarian Tumors could have helped her?

I am currently finishing PathMap Nutriton free open public discovery tool and graph. The reach may be more than the bioinformatics workbench and datasets because everyday people can ask about a food + disease and make veridical PubMed grounded discoveries.

During this build I realized that we (the world) need a “periodic table” of plant derived extracellular vesicles.

Here is the research journey:

  1. I found ginger evs contain 6-shogaol and 6-shogaol can potentially act as prophylaxis for Ebola by blocking VP40 budding, potentially via dietary modification and certainly possible that a diy aloe+water+ginger nasal gel might be feasible … NOTE Regarding my overall referencing about ginger evs, after researching cargo sizes and cribriform plate bypass needs via spermidine modification to target amyloid plaques, ginger may be preferred if mucosal adhesion is a requirement but if using the spermidine to target Mark2-eif5a/2a axis/es -not for modification, but as a beacon - for CRISPR Cas9 gene excision therapy for C9orf72 ALS / FTD, then GRAPEFRUIT EVs may be preferred due to larger cargo/payload size - and it still modifies microglial activation but via MAPK pathway/axis
  2. But Grapefruit is not always better. For example, Moxa/Borneal-modified GINGER EVs would likely be preferable for injected thermalgel delivery to Ovarian Tumor and/or Endometriosis lesions because the 6-shogaol from dried ginger /ginger (potentially could be packed in payload if need be) itself causes apoptosis of ovarian cancer cells, without chemotherapies. Grapefruit EVs do not offer the Ovarian Tumor cell apoptosis feature, and since CRISPR / mRNA siRNA etc are not delivered, the larger EV size is not needed per se.
  3. Yet again, it is not always target-benefit that provides the final deciding factor(s). Charge and binding are considerations and clearance must always be considered. For example, A hydrogel intradermal injection to deliver a therapeutic to a skin cancer lesion, or say, toenail fungus below the keratonized nailbed, likely cannot practically use ginger EVs or ginseng or other small exosome sized EVs because the lymphatic system can easily flush them away delivering off target to clearance instead of the intended target (which could be a hack to deliver to clearance systems by the way, using small evs knowing they will “flush”)

My point is - and I could go on and on talking about algael EVs, Cucumber EVs, etc. - that we need a database not only of PDEV types and payload sizes, charges, binding preferences, etc, but also modifier molecules, targets, and inherent mechanisms of the EVs themselves.

I am trying to wrap my mind around a triplicated set of data nodes where plants contain modifiers and targets, modifier nodes contain related plants and targets, and targets contain the related plants and modifiers.

I think it makes sense because you could the look at the data from a clinical needs perspective (e.g., What is the best PDEV+MOD(s)+Target+DeliveryMechanism to treat a disease at a biological axis). We could look at it from a nutritional perspective (e.g., I have disease “xyz” and get a list of foods/plants that offer potential natural gut-X axis benefit, over long term usage). And finally, from a bioinformatics angle, (e.g., What molecular pathway gaps exist in terms of modifier-build-testing that may have synergy with which PDEVs for a disease?)

I hope that makes sense. Maybe more simply put, I am looking for research needs on/about which datapoints would be helpful in a PDEV-MODIFIER-TARGET-DISEASE dataset, but also, any other feedback about building out a PDEV dataset that considers payload/cargo sizes, innate molecules, modifier binding, charge polarities, other factors, for open-public use.

Thanks in advance,

Joshua

After looking more broadly than I did last time, I think I found quite a lot of existing work that PathMap may be able to borrow from:


I think the PDEV + modifier + target + disease idea is a good direction. The one architectural change I would make early is to treat the “periodic table” / graph as a view, not as the source-of-truth data model.

Underneath it, I would keep something closer to:

Entity
  ↓
Material / Preparation
  ↓
Process
  ↓
Observation
  ↓
Assertion + Evidence
  ↓
CandidateAssembly
  ↓
MissingBridge / CompetingAssembly / NextExperiment

Then the same records can generate the three views you described:

  • clinical-needs view: disease → tissue/cell/axis → delivery constraints → PDEV/modifier/payload candidates
  • nutrition view: plant/food → native material/cargo → oral/gut interactions → pathways/diseases
  • bioinformatics view: target/pathway → supported components → missing bridges → candidate assemblies → discriminating experiment

So I would keep one canonical ginger / STAT3 / ovarian-cancer object rather than triplicating entities, and generate different projections from it.

The practical reason is that a “ginger EV” is not necessarily one stable object with one size, charge, cargo profile or activity. MISEV2023 puts substantial emphasis on source material, preprocessing, separation, characterization and storage. A useful 2025 ginger comparison also found that ultracentrifugation, sucrose-gradient UC, membrane filtration and PEG precipitation changed not only yield but stability, metabolomic composition and in-vitro activity (Ming et al., 2025).

So my low-cost default route would be:

  1. keep your PDEV / modifier / target / disease concept;
  2. add preparation/process/context as first-class data;
  3. store measurements as observations rather than universal plant properties;
  4. separate scientific assertions from their evidence;
  5. generate the periodic-table/graph UI from those records;
  6. then add a CandidateAssembly object for the PathMap-style cases where most pieces are supported but the exact integrated configuration is not.

That last object may be especially useful for PathMap.

For the datapoints you listed, this is roughly how I would store them

Most of these look safer as contextual observations than fixed properties.

Datapoint Store roughly as Important context
Plant/PDEV source species + tissue + source material cultivar/harvest state when reported
Preparation preprocessing → isolation/separation → purification → storage probably first-class
Vesicle identity reported term + normalized term + characterization evidence plant-EV nomenclature is still evolving
Size distribution + method + medium + time not just one mean diameter
Concentration/yield particles/mL or particles/g + method preparation-dependent
Charge zeta potential + medium + pH/ionic conditions + method not just “negative”
Native cargo molecule + assay + abundance if available + preparation detection ≠ functional delivery
Engineered payload cargo + loading process + efficiency/capacity + release/retention + functional assay preparation/process-specific
Modifier modifier + modification process + localization/attachment evidence + post-modification characterization entity + transformation
Binding target + assay + context distinguish association/receptor binding/internalization
Uptake cell/tissue + dose + time + assay uptake ≠ functional cargo delivery
Innate mechanism assertion + pathway + cell/species/disease context + evidence avoid universalizing context-specific effects
Delivery route route + formulation + dose + schedule critical for biodistribution
Biodistribution tissue × time × route × dose × detection method preferably a time series
Clearance PK-like observations + formulation + assay + organs/time size alone is not enough
Disease/target normalized ID + tissue/cell/phenotype context useful for joins
Evidence PMID/DOI + supporting passage/figure/table when possible + study type + context makes every edge auditable
Contradiction assertion A challenges B under context C better than calling a whole paper “contradictory”

For maximum payload size, I would avoid a hard field such as:

ginger_EV.max_payload_size = X

unless a paper really establishes such a limit. The EV-loading literature shows that cargo chemistry, vesicle composition and loading method all matter; incubation, electroporation, sonication, extrusion, freeze-thaw and transfection can have very different performance and effects on vesicle integrity (EV cargo-loading overview; reporting proposal for exogenous EV cargo loading). A systematic review of plant-derived EV drug-delivery studies likewise shows substantial variation across cargos and preparations.

I would therefore store demonstrated loading capability:

vesicle_preparation
cargo_type / identity / length or MW
loading_method
vesicle:cargo ratio
loading_efficiency / loading_capacity
free-cargo removal
post-loading size / charge
integrity
retention / release
functional-delivery assay
evidence

The same logic applies to biodistribution/clearance. A systematic review of EV biodistribution found strong size/time trends but also major methodological heterogeneity, so I would treat route, source, formulation, dose, surface properties and time as part of the observation rather than infer clearance from size alone.

I would also keep these predicates separate:

ASSOCIATES_WITH
INTERNALIZED_BY
DELIVERS_CARGO_TO
MODULATES_TARGET
CHANGES_PHENOTYPE
IMPROVES_DISEASE_ENDPOINT

because fluorescent association with a tissue is much weaker evidence than intact-vesicle delivery, functional intracellular cargo delivery, target modulation, or disease improvement.

Useful reporting standards and a small v0.1 schema

I would borrow fields rather than invent everything from scratch:

  • MISEV2023 — source material, preprocessing, separation, characterization, storage and functional claims.
  • EV-TRACK — methodological reporting turned into a queryable EV knowledgebase.
  • MIFlowCyt-EV — useful if single-EV flow data are included; separates preanalytics, preparation, controls, calibration and reporting.
  • MIRIBEL — especially useful for engineered PDEVs because it separates material characterization, biological characterization and experimental protocol.
  • ISEV Plant EV Task Force — plant-specific nomenclature/standardization is still moving, so keeping both the author’s original term and a versioned normalized mapping seems safer than freezing an ontology too early.

A small schema could begin with only:

Entity
Material
Process
Observation
Assertion
Evidence
CandidateAssembly

For example:

{
  "material_id": "MAT:GINGER:ROOT:PREP_001",
  "source": {
    "organism": "Zingiber officinale",
    "tissue": "rhizome"
  },
  "preparation": {
    "preprocessing": [],
    "separation": [],
    "purification": [],
    "storage": {}
  },
  "reported_names": [
    "ginger-derived extracellular vesicles",
    "ginger-derived exosome-like nanoparticles"
  ],
  "observations": [
    {
      "type": "particle_size_distribution",
      "method": "NTA",
      "conditions": {},
      "evidence_id": "EVID:..."
    }
  ]
}

Transformations can then be explicit:

Material A
  --[load siRNA by electroporation]-->
Material B
  --[surface modification]-->
Material C
  --[encapsulation in hydrogel]-->
Formulation D

That preserves how the engineered object actually came into existence instead of reducing everything to:

ginger_EV HAS_MODIFIER X

For normalization, existing vocabularies can carry much of the boring work: NCBI Taxonomy, NCBI Gene, ChEBI, PubChem, Mondo, Reactome, Gene Ontology, UBERON and Cell Ontology.

I would still preserve the original free-text label beside every normalized ID.

The PathMap-specific layer I would explore: CandidateAssembly

The broader search made me think the interesting unit is not merely a hidden A–B relation, but an untested integrated assembly.

PathMap dataset #134 already comes close to this in prose: it distinguishes evidence for individual GDEV / hydrogel / permeability-related components from the exact integrated configuration that remains untested.

I would make that distinction machine-readable:

CandidateAssembly
├── components
├── intended_context
├── required_commitments
├── supporting_evidence
├── challenging_evidence
├── context_compatibility
├── prior_art_status
├── missing_bridges
├── competing_assemblies
└── candidate_experiments

For example, an assembly involving a carrier + modifier + hydrogel + route + disease context might require:

C1 modifier can be attached/associated reproducibly
C2 modification preserves vesicle integrity/function
C3 hydrogel releases functional vesicles
C4 route/formulation reaches the intended compartment
C5 modifier improves the intended transport property there
C6 native/loaded therapeutic mechanism remains active
C7 integrated system has acceptable off-target effects/toxicity

The key distinction is:

support for C1–C7 individually is not support for the exact assembly.

That sounds obvious, but making it explicit is useful because compositional biomedical ideas can fail at a single interface.

It also gives PathMap a more informative output than one plausibility score:

Six commitments have direct/near-direct support; two are unresolved. The highest-risk bridge is X. Experiment Y would test that bridge and distinguish the leading alternatives.

A compact object could look like:

{
  "assembly_id": "PA:000123",
  "components": [
    {"role": "carrier", "id": "..."},
    {"role": "payload", "id": "..."},
    {"role": "modifier", "id": "..."},
    {"role": "formulation", "id": "..."}
  ],
  "context": {
    "disease": "...",
    "species": "...",
    "tissue": "...",
    "cell_type": "...",
    "route": "..."
  },
  "required_commitments": [
    {
      "id": "C1",
      "assertion": "...",
      "status": "supported|partial|unknown|challenged",
      "evidence_ids": [],
      "challenge_ids": [],
      "context_match": "high|medium|low|unknown"
    }
  ],
  "prior_art": {
    "exact": [],
    "near": [],
    "subassembly": [],
    "bridge": [],
    "searched_at": "...",
    "search_protocol": "..."
  },
  "competing_assemblies": [],
  "experiments": []
}

The graph can then be generated as a projection:

component ↔ assembly
evidence ↔ assertion
assertion ↔ commitment
commitment ↔ experiment
assembly ↔ disease/target/modifier

without making the projection itself the authoritative scientific record.

A lot of adjacent work can probably be borrowed

I found enough neighboring work that I would actively reuse concepts rather than design every layer independently.

Claims/evidence/context: INDRA and EMMAA

INDRA separates typed scientific Statements from one or more Evidence objects, including citations, source text, annotations and biological context. That is very close to what a PDEV claim graph needs.

EMMAA incrementally incorporates new literature and resolves exact/partial redundancy during assembly; its model-testing framework can rerun tests as models change. This is a useful precedent for making a PathMap discovery a living object rather than a frozen report.

Whole-vs-parts feasibility: DARPA SciFy

DARPA’s Scientific Feasibility (SciFy) program is particularly close conceptually. It decomposes capability claims into constituent verifiable parts, while explicitly noting that parts being possible does not establish feasibility of the whole: integration, logical consistency and compatibility can still matter.

That is almost exactly the failure mode a multi-component PDEV assembly has to guard against.

Mechanistic/model composition: Big Mechanism / ASKEM

DARPA’s Big Mechanism targeted reading fragmented biomedical literature and assembling causal mechanisms. ASKEM goes further into heterogeneous scientific knowledge, model composition/decomposition, simulation and traceability.

So I would not define PathMap’s distinction simply as “combining components.” A narrower and more useful target may be tracking the required commitments of a proposed experimental/therapeutic assembly.

Prior-art/contradiction search: FutureHouse / PaperQA2

FutureHouse’s science agents include a precedent-search agent intended for “Has anyone done X before?” questions. Their open-source PaperQA2 is also useful implementation precedent for iterative literature retrieval, evidence extraction and cited synthesis, and includes contradiction-search tooling.

Closed-loop discovery: Robin

The 2026 Nature paper on Robin links literature search, hypothesis generation, proposed experiments, human wet-lab work, data analysis and updated hypotheses.

So “closing the scientific loop” by itself is already becoming an occupied space. A complementary PathMap role could be to produce an auditable assembly dossier that a human researcher or later experimental agent can consume.

Competing hypotheses and discriminating measurements

A much older AI precedent is Reiter’s model-based diagnosis: keep competing explanations and choose measurements that distinguish them.

The domain is different, but the pattern maps well to:

Assembly A: modifier improves uptake
Assembly B: modifier adds nothing
Assembly C: modifier destabilizes vesicle

followed by:

Which inexpensive measurement best separates A/B/C?
Novelty, ranking, and the next experiment

I would make novelty a search state, not a Boolean scientific property.

Instead of:

novel = true

something like:

EXACT: no match found
NEAR: 3 matches
SUBASSEMBLY: 12 matches
BRIDGE: partial precedent
searched_at: 2026-08-28
sources: [...]
query_families: [...]
coverage_limitations: [...]

Then the public wording can be:

No exact prior art was found under this search protocol as of DATE.

That is stronger scientifically than “this has never been done,” because synonymy, full text, patents, trial registries, conference material and database coverage can all defeat a simple novelty search.

This caution also has precedent in LBD evaluation. A 2023 Bioinformatics analysis describes weaknesses in evaluating literature-based discoveries using a small set of famous retrospective cases or later co-occurrence as a gold standard.

For ranking, I would therefore look at:

support_directness
context_compatibility
contradiction_severity
evidence_independence
missing_bridge_count
missing_bridge_severity
prior_art_distance
experimental_tractability

and keep bridge count separate from bridge severity. One unresolved question such as “does this reach the relevant spinal motor neurons?” may be much more serious than three cheap formulation-compatibility questions.

For experiments, I would rank not only “what tests the hypothesis?” but:

Which experiment removes the most uncertainty or best distinguishes the leading assemblies?

A candidate experiment could carry:

tests_commitments
discriminates_assemblies
expected_outcomes_by_hypothesis
cost_class
time_class
technical_complexity

Then a cheap integrity/loading assay can outrank an animal study if it can already falsify the whole assembly.

Finally, I would benchmark the system before scaling it:

  1. known-prior-art test — can it avoid calling established assemblies new?
  2. retrospective bridge test — using literature only up to year N, can it identify bridges later experimentally supported?
  3. adversarial assembly test — all components are individually supported, but their integration fails because of route/context/loading/formulation/etc.
  4. context-vs-contradiction test — can it distinguish genuine conflict from different species/cells/doses/preparations/endpoints?
  5. experiment-selection test — does its top experiment actually discriminate alternatives better than a generic next step?

Those seem more informative than simply counting generated hypotheses.

A small v0.1 would probably be enough to learn whether this structure is useful:

  1. manually curate 20–50 well-documented PDEV preparations;
  2. preserve preparation/process/context/evidence;
  3. generate the three views from one canonical dataset;
  4. add 5–10 CandidateAssemblies with only a few unresolved bridges;
  5. run exact/near/subassembly/bridge precedent searches;
  6. see whether PathMap can clearly answer:
What is directly known?
What is inferred?
What is challenged?
What exact combination was searched?
What bridge is still missing?
What cheap experiment would resolve the most uncertainty?

That would keep the project very close to the direction you already seem to be exploring rather than turning it into a different system.

The “periodic table” idea can remain the friendly interface. The main change would be to give it a preparation- and evidence-aware source of truth underneath.

And the PathMap-specific design target I would personally explore is this:

a near-complete experimental assembly whose individual commitments, evidence, context, contradictions, prior-art search, missing bridges, competing variants and next discriminating experiment are tracked together.

I would be cautious about claiming that exact combination is unprecedented—the neighboring fields are broad and moving quickly—but it looks like a useful object even if every individual piece has prior art elsewhere.

Thank you. After thinking about this for several days, I think that it would be “over my head” to start at a structured data graph, for many of the cautious reasons you gave.

What I decided to do moving forward is set up a free ai no-barriers (just recaptcha, no emails etc) open public system that lets anyone enter a Food, Plant, or Dietary Molecule PLUS a Pathology, Symptom, or Condition and then use the logic from PathMap Core to simply make a Swanson’s Style discovery from a mix of PubMed+OpenAlex abstract set the system will use AI to suggest Search Booleans and then pull API for abstracts and build and veridicality enforcement queries.

Interestingly (to me) is we get:

User input tags/phrases which then trigger AI generated search terms (booleans) which then produce the Abstracts which have MeSH tags which then is narrowed to 2-5 abstracts for the discovery and also a set of AI tags that pertain to the overall summary of how item A (food/plant/molecule) relates to the stated pathology. The has to be some kind of tags attractor in that data flow so I will likely provide monthly datasets for free once compiled if anyone can use it.

This is the basic user flow but it goes on in background from there.

I am releasing the PathMap Admin Version (minus api key of course) as Apache 2.0 and “nuking” any potential future patent enforceability (planned for Tuesday).

I will then set up my laptop to do pathMap Core builds about the initial pairing as well as the discovery the user spawned and integrate linking into the original post.

It gets to PDEVs DB stuff from here…

I think PubChem Data and SMILES strings are readily available for many of the items that can be extra yes from the created datasets where first, the system programatically fills in and creates a flat JSON database of known data about the involved molecules, and AI can search for PDEV-related information if it cannot be derived from pubChem data.

so, instead of a DB that tries to map theoretical PDEV usage, it instead looks for known therapeutics and contraindications regarding A versus B that can be used afterwards by a system+AI to look for potential synergies between states of a food (e.g., dry ginger is more of a 6-shogaol source whereas ginger juice is more of a 10-shogaol and fried is something else {cant remember TDEV or something}) and specifically identifies gaps and know data.

An example might be where we need to know the payload size of a PDEV or cargo size of a molecule and the data is identified. Next, we want to know about its hydrophilic/lipophilic/electrophilic properties and PubChem SMILES data typically can fill the blanks. As a fallback AI can look at note the missing PubChem data versus available data, or, if still missing, flag as a gap.

Knowing Cargo/Payload sizes is more than just packaging. Small exosomes can flush from skin in intradermal injection (e.g. ginseng/ginger) and would not be likely preferred to treat a skin cancer via hydrogel, but alternatively, could be used to deliver therapeutics locally to lymph system, eg, in the case of breast cancer)… Grapefruit EVs may be ideal for some uses but also has a know contraindication, etc.

But… I think the data will have “gravity” for some specific PDEVs as being “potentially more ideal” than others, when delivered via specified route, for a specific pathology, symptom, etc.

I initially really liked idea Ginger EVs via intranasal and modifying with spermidine to deliver CRISPR Cas9 for a potential C9orf72 ALS/FTD targeted nanomed system.

It may be a good idea but Grapefruit EVs can deliver a larger payload (I havent looked at “Spermidine-modified GfEVs” yet as a potential). This is why the open public seeded discovery system that then feeds the AI for what to look at (admittedly, not comprehensive, but for me, this is a starting point.to build upon) seems like the right direction to me.

I will mark this as the solution for this thread and when the Apache 2.0 PathMap code is released I will post in the appropriate thread/forum. As for the Nutrition page for PathMap, it will be added this seek and the 144 studies archived as my personal use of the software/demo blog. It makes sense to me to do something that most people can relate to - food and illness - and allow the discovery engine + php/API and AI resolve data that is known and what is not yet discovered by the system.

The software is now open source and may be downloaded at PathMap V2.0 Open Source Admin/Developer Version - Universal AI Workbench by PathMap.org

All 144 publishings at pathmap(.org) were made using this software.

Thank you all for your inputs on the literature based discoveries about ALS and Breast Cancer, PDEVs, etc. It is Apache 2.0 License now so have at it.