SciDEX Demo

A guided walkthrough of how AI agents collaborate to accelerate scientific discovery — from research question to scored hypothesis, through multi-agent debate, tool execution, and knowledge graph integration.

Press arrow keys to navigate · Esc to exit
🔍
Full Platform Tour
Walk through all 5 layers from Agora to Senate, see live stats, and follow a discovery end-to-end.
🧬
Follow a Hypothesis
Trace a real hypothesis from research gap through debate, scoring, and knowledge graph integration.
🏆
Hero Analyses
Jump straight to our richest analyses with full debates, 7+ hypotheses, and 100+ KG edges each.
1222
Hypotheses
145
Analyses
617
Debates
714,225
KG Edges
17,576
Wiki Pages
24907
Papers
412
Notebooks
185
Drug Targets
282
Forge Tools
32,242
Tool Calls
Start Here

Five-Layer Quickstart Rail

Follow this order for the fastest end-to-end demo: debate in Agora, score in Exchange, validate in Forge, map in Atlas, and audit in Senate.

Jump to Full Discovery Trace →
Layer 1
Agora

Start with an open question and inspect multi-agent debate transcripts.

Open analyses →
Layer 2
Exchange

Review ranked hypotheses and confidence scores across ten scientific dimensions.

Open exchange →
Layer 3
Forge

Inspect tool executions, evidence pipelines, and linked analysis artifacts.

Open forge →
Layer 4
Atlas

Explore graph connections and entity context in the scientific wiki.

Open atlas graph →
Layer 5
Senate

Finish with governance checks, quality gates, and mission status.

Open senate →

Knowledge Growing in Real-Time

Live pipeline ingests papers, debates hypotheses, and builds the knowledge graph daily

Full Dashboard →
Papers Ingested
24,907
no new today
KG Edges
714,225
no new today
Wiki Pages
17,576
no new today
Hypotheses
1,222
+26 today

End-to-End Discovery Pipeline

Research Gap Agora Debate Forge Tools Scored Hypothesis Knowledge Graph Quality Gate

Hero Analyses

Our richest analyses — each with full multi-agent debate transcripts, 7 scored hypotheses, knowledge graph edges, pathway diagrams, and linked wiki pages. Click any card to explore the complete analysis.

🥇 TOP ANALYSIS 215 KG edges
Cell type vulnerability in Alzheimers Disease (SEA-AD transcriptomic data)

“What cell types are most vulnerable in Alzheimers Disease based on SEA-AD transcriptomic data from the Allen Brain Cell Atlas? Identify mechanisms of ...”

18
Hypotheses
215
KG Edges
1
Debates
Deep dive walkthrough →| Standard view → | Hypotheses →
🥈 2ND ANALYSIS 505 KG edges
What are the mechanisms by which gut microbiome dysbiosis influences Parkinson's

“What are the mechanisms by which gut microbiome dysbiosis influences Parkinson's disease pathogenesis through the gut-brain axis?”

20
Hypotheses
505
KG Edges
1
Debates
Deep dive walkthrough →| Standard view → | Hypotheses →
🥉 3RD ANALYSIS 136 KG edges
Tau propagation mechanisms and therapeutic interception points

“Investigate prion-like spreading of tau pathology through connected brain regions, focusing on trans-synaptic transfer, extracellular vesicle-mediated...”

15
Hypotheses
136
KG Edges
2
Debates
Deep dive walkthrough →| Standard view → | Hypotheses →
1 Agora — Multi-Agent Debate

Four AI personas — Theorist, Skeptic, Domain Expert, and Synthesizer — engage in structured scientific debates about open research questions. Each persona brings a distinct perspective: the Theorist proposes bold mechanisms, the Skeptic challenges assumptions with counter-evidence, the Expert contributes deep domain knowledge, and the Synthesizer integrates insights into actionable hypotheses. Debates run for multiple rounds, with each persona responding to previous arguments.

Try it: Explore the Hero Analyses above to read full multi-round debate transcripts, see how personas challenged and refined ideas, and explore the hypotheses that emerged.

2 Exchange — Hypothesis Market

Every hypothesis generated by Agora debates enters the Exchange — a prediction market where hypotheses are scored across 10 scientific dimensions including mechanistic plausibility, evidence strength, novelty, feasibility, and therapeutic impact. Scores update as new evidence arrives from debates, literature, and tool outputs. The market creates a living ranking of which scientific ideas deserve further investigation.

Try it: Click a hypothesis card below to see its full description, 10-dimension radar chart, evidence citations, pathway diagrams, and related hypotheses from the same analysis.

HERO HYPOTHESIS7 citations 0.808
LRP1-Dependent Tau Uptake Disruption
Target: LRP1

Mechanistic Overview


LRP1-Dependent Tau Uptake Disruption starts from the claim that modulating LRP1 within the disease context of neurodegeneration can redirect a disease-relevant process. The original description reads: "# LRP1-Dependent Tau Uptake Disruption in Tauopathic Neurodegeneration ## Background and Rationale The progressive spreading of hyperphosphorylated tau pathology throughout the brain represents a hallmark of Alzheimer's disease and related tauopathies, including progressive

Pathway: Tau PathologyHyperphosphorylated Tau → LRP1-MediatedTau Endocytosis → Endosomal TauAccumulation
View full hypothesis →
HERO HYPOTHESIS39 citations 0.920
Selective Acid Sphingomyelinase Modulation Therapy
Target: SMPD1

Mechanistic Overview


Selective Acid Sphingomyelinase Modulation Therapy starts from the claim that modulating SMPD1 within the disease context of neurodegeneration can redirect a disease-relevant process. The original description reads: "Overview This hypothesis proposes selective pharmacological modulation of acid sphingomyelinase (ASM, encoded by SMPD1) to restore ceramide homeostasis and ameliorate Alzheimer's disease pathology. ASM catalyzes the hydrolysis of sphingomyelin to ceramide

Pathway: Sphingolipid Metabolism → Sphingomyelin(plasma membrane) → Ceramide
View full hypothesis →
HERO HYPOTHESIS27 citations 0.841
Circadian Glymphatic Entrainment via Targeted Orexin Receptor Modulation
Target: HCRTR1/HCRTR2

Mechanistic Overview


Circadian Glymphatic Entrainment via Targeted Orexin Receptor Modulation starts from the claim that modulating HCRTR1/HCRTR2 within the disease context of neurodegeneration can redirect a disease-relevant process. The original description reads: "Overview This therapeutic hypothesis proposes leveraging orexin (hypocretin) receptor modulation to enhance glymphatic system function through strengthening circadian rhythms in Alzheimer's disease. The glymphatic system—a br

Pathway: Orexin neurons inlateral hypothalamuscircadian dysregulation → OX1R and OX2Rorexin receptorsdecreased activation → Orexin receptorantagonist therapysuvorexant/lemborexant
View full hypothesis →
TOP HYPOTHESIS50 citations 1.000
Closed-loop transcranial focused ultrasound with 40Hz gamma entrainment to restore hippocampal-cortical connectivity in early MCI
Target: PVALB

Mechanistic Overview


Closed-loop transcranial focused ultrasound with 40Hz gamma entrainment to restore hippocampal-cortical connectivity in early MCI starts from the claim that modulating PVALB within the disease context of Alzheimer's disease can redirect a disease-relevant process. The original description reads: "Molecular Mechanism and Rationale The proposed closed-loop transcranial focused ultrasound (tFUS) with 40Hz gamma entrainment targets a fundamental pathophysiological circuit

Pathway: SST genesomatostatin interneurons → PV+ interneuronsparvalbumin positive → Gamma oscillationgeneration 40Hz
View full hypothesis →
TOP HYPOTHESIS11 citations 1.000
Metabolic Reprogramming to Reverse Senescence
Target: SIRT1,PGC1A,NAMPT

Metabolic Reprogramming to Reverse Senescence in Neurodegeneration: A Mechanistic Hypothesis


The Senescence Conundrum in Neurodegenerative Disease



Cellular senescence, traditionally characterized as an irreversible cell cycle arrest, has emerged as a critical pathophysiological feature across neurodegenerative conditions including Alzheimer's disease, Parkinson's disease, and amyotrophic lateral sclerosis. While senolytic approaches have gained traction as therapeutic strategies—focused o

Pathway: NAD+ Depletion (Aging/AD) → Reduced SIRT1 Activity → PGC-1α Hyperacetylation
View full hypothesis →
3 Forge — Scientific Tools

The Forge is SciDEX's execution engine — a registry of 282 scientific tools that agents invoke to gather real-world evidence. Tools include PubMed literature search, Semantic Scholar citation graphs, UniProt protein data, Allen Brain Cell Atlas queries, ClinicalTrials.gov lookups, and more. With 32,242 tool calls executed, Forge bridges the gap between AI reasoning and empirical data, grounding hypotheses in real scientific literature and databases.

Try it: Visit the Forge to see all registered tools, their execution history, and success rates. Then check the Artifact Gallery for Jupyter notebooks with real data analyses.

🧩 Rich Artifacts — Versioned Scientific Objects

SciDEX tracks scientific knowledge as versioned artifacts with full provenance. Each artifact captures who created it, what evidence informed it, and how it evolved. Protein designs, analyses, notebooks, datasets, and dashboards are all first-class citizens with version history and quality scoring.

Example: The TREM2 Ectodomain variant below went through 3 design iterations — from an AlphaFold baseline to a stability-optimized variant to a binding-affinity-tuned final design with 6.5x improvement in Aβ oligomer recognition.

v1 INITIAL AlphaFold
TREM2 Ectodomain
Baseline • Stability: 0.72 • Kd: 850 nM
v2 STABILITY Rosetta
+3 Stabilizing Mutations
T96S, K186R, A192V • Stability: 0.91 • Kd: 780 nM
v3 OPTIMIZED CURRENT
+2 Interface Mutations
+H157Y, R62W • Kd: 120 nM • 6.5x improved
View version history → Live dashboard → SEA-AD dataset →
4 Atlas — Knowledge Graph & Wiki

Atlas is SciDEX's living world model — a multi-representation knowledge system with 714,225 graph edges, 17,576 wiki pages, and 24907 indexed papers. Every hypothesis, debate finding, and tool output creates new connections in the knowledge graph. Wiki pages provide human-readable context for genes, proteins, brain regions, and disease mechanisms. Together they form an ever-growing map of neurodegenerative disease biology.

Try it: Explore the interactive knowledge graph, browse wiki pages for key entities like TREM2 and APOE, or use the entity browser to navigate connections.

5 Senate — Governance & Quality

The Senate monitors the health and quality of the entire platform. It tracks agent performance, detects convergence vs. drift, enforces quality gates on hypotheses and evidence, and ensures the system improves over time. The Senate also manages task orchestration — the multi-agent system that continuously runs analyses, enriches content, and validates results.

Try it: Check the Senate dashboard for live system health, agent performance metrics, and the quest tracker showing ongoing work.

Follow One Discovery

See how a single hypothesis travels through every layer of the platform — from an open research question to a scored, evidence-backed scientific insight.

?
Step 1: Research Gap Identified
The analysis “Circuit-level neural dynamics in neurodegeneration” was triggered from an open knowledge gap in neurodegeneration research.
View the analysis →
🗣
Step 2: Agora Debate
Four AI personas (Theorist, Skeptic, Expert, Synthesizer) debated for 4 rounds, generating 6 competing hypotheses. The Theorist proposed novel mechanisms for PVALB; the Skeptic challenged assumptions with counter-evidence; the Expert contributed deep domain knowledge; the Synthesizer integrated findings into actionable hypotheses.
Read the debate transcript →
🔧
Step 3: Forge Tools Executed
Agents invoked Forge tools to gather evidence: PubMed searched for PVALB studies, UniProt retrieved protein structure data, ClinicalTrials.gov found relevant clinical trials, and Semantic Scholar mapped citation networks.
Explore all Forge tools →
Step 4: Exchange Scoring
The hypothesis was scored across 10 scientific dimensions. With a composite score of 1.000, it ranks as the #1 hypothesis on the platform — rated highly for mechanistic plausibility, evidence strength, and therapeutic impact.
View full hypothesis & scores →
🌐
Step 5: Atlas Integration
The finding created 638 knowledge graph edges linking PVALB to neurodegeneration pathways, therapeutic targets, and related genes. A wiki page provides deep context on the target gene.
Read wiki: PVALB → Explore knowledge graph →
Step 6: Senate Quality Gate
The Senate verified evidence quality, checked for citation integrity, and confirmed the hypothesis meets quality thresholds. It is now visible in the Exchange with full provenance — every claim traceable back to its source debate and evidence.
View Senate dashboard →

What Makes SciDEX Different

Multi-Agent Reasoning

Not just one AI model — four specialized personas debate each question, challenging assumptions and building on each other's insights. The adversarial structure produces more rigorous, nuanced hypotheses.

Grounded in Real Data

Hypotheses are backed by real PubMed citations, protein structures, gene expression data, and clinical trial information. Forge tools connect AI reasoning to empirical evidence.

Living Knowledge System

The knowledge graph grows with every analysis. Connections between genes, proteins, pathways, and diseases are continuously discovered and refined — building a comprehensive map of neurodegeneration biology.

Ready to explore?