Building Synapedia.com — A Knowledge Graph for Psychoactive Substances

Florian Lux — Wed, 11 Mar 2026 04:01:17 +0000

Most information about psychoactive substances is scattered across the internet.

Some of it lives in scientific papers.

Some of it lives in harm-reduction communities.

Some of it is buried in old forum posts.

Some of it exists as fragmented experience reports.

Some of it is hidden behind technical pharmacology language that most people will never read.

The problem is not that information does not exist.

The problem is that it is rarely connected.

So I started building Synapedia.com: a structured knowledge platform for psychoactive substances, interactions, pharmacology and harm reduction.

Not as a drug wiki.

Not as a forum.

Not as a recreational guide.

But as a searchable, interconnected knowledge system.

A place where substances are not just isolated pages, but nodes in a larger graph of effects, mechanisms, risks, receptors, combinations, timelines and real-world context.

Psychoactive substances should not be understood as isolated entries.

They should be understood as part of a network.

The problem: substance information is fragmented

If someone wants to understand a psychoactive substance, they usually have to jump between completely different kinds of sources:

scientific papers
pharmacology databases
community reports
harm-reduction websites
old forum discussions
anecdotal timelines
medical pages
scattered Reddit threads
substance-specific wikis

Each source may contain one useful piece of the puzzle.

But the actual questions people have are usually relational:

What does this substance interact with?
Which receptors or transporters are involved?
Which effects are commonly reported?
How long does it last?
Which combinations are dangerous?
Which substances are similar?
Which risks are specific, and which are class-based?
What is known, and what is still uncertain?

Traditional articles are not great at answering these questions.

Because the relationships are often more important than the isolated facts.

That is where Synapedia begins.

The idea: substances as a graph, not a list

Most websites treat substances like encyclopedia entries.

Synapedia treats them as connected entities.

A substance can be connected to:

effects
receptors
transporters
substance classes
mechanisms
risks
interactions
reports
timelines
related compounds
educational articles
comparison pages

That structure makes it possible to explore psychoactive substances in a more intuitive way.

You can start with a compound, move to its effects, compare it with another substance, check interaction risks, read structured reports, and then explore the pharmacological mechanisms behind it.

This is the core idea:

A substance page should not be a dead encyclopedia entry.

It should be a living node in a larger knowledge graph.

What I built

Synapedia currently combines several product surfaces into one platform.

It is not just one page type.

It is a system of connected tools, structured content and searchable relationships.

1. Substance profiles

Each substance has a structured profile designed to answer the most important questions quickly:

What is it?
How does it feel?
How long does it last?
What are the main risks?
Which interactions matter?
What mechanisms are involved?
What should users be careful about?

The goal is not to create long unreadable pages.

The goal is to make the first screen useful immediately, while still allowing deeper exploration further down the page.

A good substance page should feel like a dashboard:

quick orientation at the top
effects and timeline early
risk signals visible
interaction warnings easy to find
mechanisms and sources available for deeper reading

This is especially important because people searching for substance information are often not casually browsing.

They may be trying to understand something urgent, confusing or risky.

2. Interaction checker

One of the most important parts of Synapedia is the interaction system.

Many of the highest-risk situations do not come from one substance alone.

They come from combinations.

Alcohol plus benzodiazepines.

Opioids plus sedatives.

Serotonergic substances plus certain antidepressants.

Stimulants plus other stimulants.

Research chemicals plus unknown potency.

The interaction checker is designed to make these risks more visible.

It is not a medical diagnosis tool.

It is a harm-reduction interface.

The purpose is to surface warnings, mechanisms and risk categories before users rely on guesswork.

This is also one of the hardest areas, because interaction data quality varies a lot.

Some combinations are well documented. Others are inferred from pharmacology or substance class.

Synapedia tries to make that uncertainty visible instead of pretending every data point has the same confidence.

3. Compare pages

People do not only search for one substance.

They compare.

MDMA vs 2C-B.

Amphetamine vs methylphenidate.

Ketamine vs DXM.

LSD vs psilocybin.

Kratom vs O-DSMT.

Comparison is one of the most natural ways humans understand complex things.

So I built compare pages that place two substances next to each other and show differences in:

subjective effects
duration
risk profile
mechanisms
interaction concerns
substance class
practical context

This turns Synapedia from a passive database into an exploratory tool.

Instead of reading isolated pages, users can understand relationships.

4. Knowledge graph explorer

The knowledge graph is one of the parts I am most excited about.

It visualizes connections between substances, receptors, effects, mechanisms, articles and risks.

This matters because pharmacology is relational by nature.

Two substances may look unrelated by name but share receptor activity.

Two compounds may be in the same class but differ strongly in duration or risk profile.

A subjective effect may appear across multiple substance families for completely different mechanistic reasons.

A graph makes these relationships visible.

And once you start seeing substances as a network, the entire topic becomes easier to understand.

5. Knowledge articles

Structured data alone is not enough.

Some topics need explanation.

That is why Synapedia also includes knowledge articles about substance classes, receptor systems, interactions, withdrawal, research chemicals, risk contexts and harm-reduction principles.

The goal is to connect short, searchable substance profiles with deeper educational material.

A user might start with a substance page, move to an interaction warning, open a receptor article, compare two substances and then explore the graph.

That is the product loop.

Not isolated pages.

Connected knowledge.

The architecture

Synapedia is built with a modern web stack:

Next.js
React
TypeScript
Tailwind CSS
Supabase
PostgreSQL
structured JSON data
automated enrichment workflows
sitemap generation
bilingual routing
schema.org structured data
static fallbacks for resilience

The platform is designed around one principle:

Public pages should work as a fast, structured knowledge layer — even when the underlying data sources evolve.

At a high level, the system looks like this:

Substance data
   ↓
Normalization and enrichment
   ↓
Structured profiles
   ↓
Effects, receptors, mechanisms and interactions
   ↓
Search, comparison, graph and article surfaces
   ↓
SEO-ready public knowledge pages

The hard part is not just building pages.

The hard part is building a system where each piece of knowledge can support multiple surfaces:

a substance profile
a compare page
an interaction result
a graph node
a related article
a structured report
a search result

That is where Synapedia starts to feel less like a website and more like a knowledge engine.

Technical challenges

Building this has been much more complex than a normal content site.

Some of the main challenges:

Data quality

Psychoactive substance data is uneven.

Some compounds are well studied.

Others are obscure research chemicals with limited public information.

Some interaction risks are documented.

Others are inferred from class, mechanism or partial evidence.

So the platform needs to separate:

known information
inferred information
class-based templates
user-facing summaries
source-backed content
uncertain or incomplete areas

This is especially important for harm reduction.

A system like this should never pretend that all data has equal confidence.

SEO without thin content

Programmatic SEO can be powerful.

But it can also become dangerous.

If you generate thousands of weak pages, Google will eventually treat the site as low quality.

So Synapedia uses readiness logic to decide which pages should be indexable and which should stay out of the search index until they have enough depth.

The goal is not to publish the most pages.

The goal is to publish the right pages.

User experience under pressure

Substance information is not like a recipe blog.

A user may arrive because they are worried about an interaction, trying to understand an effect, or comparing two compounds before making a risky assumption.

That means the interface has to prioritize:

clarity
fast scanning
mobile readability
visible warnings
structured sections
no unnecessary friction
no false certainty

Design is not decoration here.

Design is part of harm reduction.

Trust

This is a sensitive topic.

If a platform discusses psychoactive substances, risks, interactions and withdrawal, it needs to be careful.

Synapedia is not medical advice.

It does not replace doctors, emergency services or addiction support.

But it should still aim for strong trust signals:

clear disclaimers
source visibility
review status
evidence levels
transparent uncertainty
no sourcing or buying information
no glorification
strong harm-reduction framing

Trust is not just a UI component.

It is part of the product.

Why this matters

People are already searching for this information.

They are searching before, during and after experiences.

They are searching after reading a forum post.

They are searching when they are unsure about a combination.

They are searching when they do not understand what happened.

They are searching when official information is too abstract or too judgmental.

If the information they find is chaotic, incomplete or misleading, risk increases.

A better knowledge system will not solve every problem.

But it can make dangerous uncertainty smaller.

That is the purpose of Synapedia.

What makes Synapedia different

The most important difference is that Synapedia is not just a collection of pages.

It is a connected system.

A substance page can lead to:

related mechanisms
similar compounds
risk categories
interaction warnings
comparison pages
educational articles
graph exploration

This creates a feedback loop:

More structure makes better navigation possible.

Better navigation makes the knowledge more useful.

More useful pages create stronger search surfaces.

Stronger search surfaces bring more feedback and more data gaps to fix.

That is the long-term moat.

Current focus

Right now, I am focused on improving:

substance page quality
interaction depth
trust signals
internal linking
knowledge graph visibility
bilingual SEO
structured report parsing
mobile-first readability
evidence and source transparency

Synapedia is still evolving quickly.

But the foundation is there:

A structured public knowledge platform for one of the most fragmented and misunderstood areas of health, neuroscience and culture.

The vision

The long-term vision is simple:

Build the clearest, most useful and most responsible open knowledge system for psychoactive substances.

Not a place that glamorizes use.

Not a place that hides risk.

Not a place that reduces everything to fear.

But a platform that treats people like adults and gives them structured, understandable, evidence-oriented information.

Because when people are going to search anyway, the quality of what they find matters.

Synapedia.com is my attempt to build something better.

A map for a difficult topic.

A graph instead of a pile of disconnected facts.

A harm-reduction platform built for the internet we actually have.

Explore Synapedia

You can explore the project here:

Synapedia.com

Useful starting points:

Feedback from developers, researchers, designers and harm-reduction communities is very welcome.

DEV Community: Florian Lux