Explore the Gut-Brain Connection
Upload your microbiome dataset or explore our demo data to discover correlations between gut bacteria abundance and mental health markers.
🔬 About This Dataset
This visualization uses research-based synthetic data generated from published scientific findings. The correlation patterns reflect real associations found in peer-reviewed studies linking gut bacteria to mental health.
Data Source
Synthetic data modeled on findings from Valles-Colomer et al. (2019) Nature Microbiology and Radjabzadeh et al. (2022) Nature Communications.
What We're Showing
Which bacteria are associated with better mental health vs higher symptoms based on published research.
Research Highlights
Coprococcus and Dialister depleted in depression. Eggerthella elevated in depression. Faecalibacterium linked to quality of life.
Dataset Overview
Key statistics from your microbiome + mental health data
Mental Health Score Distribution
How many subjects fall into each anxiety/depression range? Higher scores = more symptoms.
Top 10 Bacteria by Abundance
The most common gut bacteria across all subjects (% of total microbiome).
The Gut-Brain Connection
Understanding how your microbiome influences mental health
🧠 What is the Gut-Brain Axis?
Your gut and brain are in constant communication through a network called the gut-brain axis. This bidirectional highway includes:
🔌 Vagus Nerve
The longest cranial nerve directly connects your gut to your brainstem, sending signals in both directions.
🧪 Neurotransmitters
~95% of your body's serotonin is made in the gut. Gut bacteria produce GABA, dopamine, and other mood chemicals.
🛡️ Immune System
70% of immune cells live in your gut. Inflammation from gut issues can affect brain function.
🧬 Metabolites
Bacteria produce short-chain fatty acids (SCFAs) that cross the blood-brain barrier and influence mood.
😌 "Psychobiotic" Bacteria
Research suggests these species may have positive effects on mental health:
Shown to reduce anxiety and depression-like behavior in animal studies. Produces GABA.
Human trials show reduced cortisol levels and improved stress coping.
Major butyrate producer. Low levels linked to depression and IBD.
⚠️ Bacteria to Watch
Elevated levels of these may correlate with mental health symptoms:
Overgrowth linked to gut inflammation and systemic effects. Often rises after antibiotic use.
Thrives on saturated fat. Produces hydrogen sulfide, linked to inflammation.
Elevated in some depression studies. Produces endotoxins that may trigger inflammation.
🧪 How to Get Your Microbiome Tested
Several consumer services offer gut microbiome testing. Here's what's available:
RNA-based analysis. Provides food recommendations. Reports are user-friendly but data export is limited.
Combines microbiome, blood sugar, and blood fat testing. Personalized nutrition program included.
16S sequencing with raw data export (CSV). Good for researchers and data enthusiasts.
Includes Onegevity analysis. Clinician-friendly reports with actionable insights.
📚 Research This Tool Is Based On
Primary Sources:
• Valles-Colomer M, et al. (2019) "The neuroactive potential of the human gut microbiota in quality of life and depression." Nature Microbiology, 4(4):623-632. doi:10.1038/s41564-018-0337-x
• Radjabzadeh D, et al. (2022) "Gut microbiome-wide association study of depressive symptoms." Nature Communications, 13:7128. doi:10.1038/s41467-022-34502-3
• Simpson CA, et al. (2021) "The gut microbiota in anxiety and depression - A systematic review." Clinical Psychology Review, 83:101943.
Disclaimer: This tool uses synthetic data modeled on published research findings. It is for educational purposes only and should not be used for clinical decisions.
My Microbiome Profile
Enter your test results to see personalized insights
📝 Enter Your Data
Have results from a microbiome test? Enter the abundance percentages for key bacteria below. Don't have exact numbers? Use the sliders to estimate based on whether your report showed "low", "normal", or "high" levels.
😌 Beneficial Bacteria
Higher is generally better for these species
⚠️ Watch These Levels
Lower is generally better for these species
What-If Simulator
Explore how different microbiome compositions might relate to mental health
🎮 How This Works
This simulator lets you explore hypothetical scenarios. Adjust the bacteria levels and see how the predicted anxiety risk changes based on patterns from research data. This is for educational exploration only — real health decisions should involve healthcare providers.
🦠 Adjust Bacteria Levels
Predicted Anxiety Risk
Based on correlations in research data
Contributing Factors
💡 Insight
Adjust the sliders to see how different bacterial compositions might affect mental health risk based on research patterns.
📊 Understanding Correlations
A correlation measures how two things move together. A value of +1.0 means "when this bacteria is high, anxiety is high." A value of -1.0 means "when this bacteria is high, anxiety is low." Values near 0 mean no relationship.
🧠 Key Insight
Look for bacteria with strong negative correlations — these may be "protective" species worth cultivating through diet.
⚠️ Correlation ≠ Causation
Just because bacteria X correlates with anxiety doesn't mean it causes anxiety. The relationship could be reversed or driven by a third factor.
Gut-Brain Correlations
Which bacteria are associated with higher or lower mental health scores?
Anxiety Correlations
Red = positive (more bacteria → more anxiety). Teal = negative (more bacteria → less anxiety).
Depression Correlations
Same logic: negative correlations suggest protective effects.
Full Correlation Table
★ indicates potentially significant correlations (|r| > 0.3). Sorted by anxiety correlation strength.
| Species | Anxiety Correlation | Depression Correlation | Significant? |
|---|
🎯 What is PCA & Clustering?
PCA (Principal Component Analysis) takes 25 bacteria measurements and squishes them into 2 dimensions so we can visualize them on a plot. Each dot is a person, positioned by their overall microbiome "fingerprint."
K-Means Clustering groups similar microbiome profiles together. If anxious people cluster separately from calm people, it suggests their gut bacteria are systematically different.
🔍 What to Look For
Do the teal dots (low anxiety) cluster apart from coral dots (high anxiety)? If yes, microbiome composition may be predictive of mental health.
PCA & Clustering
Visualizing microbiome patterns across subjects
PCA - Colored by Anxiety Level
Each dot = one person. Position = their microbiome profile. Color = their anxiety level.
K-Means Clusters (k=3)
Algorithm-detected groups based purely on microbiome similarity (ignoring mental health scores).
Variance Explained by Principal Components
PC1 captures the most variation in the data. The first 2-3 components often capture the "big picture" patterns.
🔮 Predictive Modeling
Can we predict someone's anxiety level just from their gut bacteria? This simple model uses the bacteria most strongly correlated with anxiety to make a prediction. Enter abundance values (or use the defaults) and see what the model predicts.
⚗️ How It Works
The model weights each bacteria by its correlation with anxiety. High levels of "protective" bacteria lower the predicted anxiety; high levels of "risky" bacteria raise it.
Anxiety Predictor
Enter microbiome values to predict anxiety classification
Microbiome-Based Prediction
Adjust the abundance values for key bacteria (0-100% scale). These are the 6 most predictive species.
Feature Importance
Which bacteria have the strongest relationship with anxiety? Higher bars = more predictive power (regardless of direction).
About GutMind Explorer
Purpose, methodology, and limitations
🎯 What This Platform Does
GutMind Explorer is an interactive educational tool that makes cutting-edge microbiome-mental health research accessible and explorable. Instead of reading dense academic papers, users can visually explore the relationships between gut bacteria and mental health outcomes through interactive charts, ML predictions, and real-time simulations.
📊 Explore Research Patterns
Visualize correlations, clusters, and distributions from data modeled on findings published in Nature Microbiology, Nature Communications, and other peer-reviewed journals.
🤖 ML-Powered Predictions
Enter your gut bacteria levels and receive an anxiety risk assessment from a real ensemble model (Random Forest + Gradient Boosting) — the same techniques used in research.
🎮 Simulate Interventions
Adjust bacteria levels with sliders and watch risk scores update in real-time. See how dietary changes that shift specific bacteria could affect mental health markers.
📁 Bring Your Own Data
Upload results from commercial microbiome tests (Viome, Thryve, etc.) to get a personalized profile comparison against research-based population data.
🔬 What's Real vs. What's Simulated
Full transparency — here's a precise breakdown of what comes from science and what we generated:
✅ REAL: Which bacteria go up or down
The direction of every correlation comes directly from published research. When you see Coprococcus as "protective" and Eggerthella as "risk-associated" — that reflects what scientists actually found in real patient studies (Valles-Colomer 2019, Radjabzadeh 2022, Jiang 2015, Simpson 2021).
✅ REAL: The clinical scales
Mental health scores use GAD-7 (0–21) for anxiety and PHQ-9 (0–27) for depression — the same validated instruments your doctor uses. The severity cutoffs (minimal, mild, moderate, severe) are standard clinical thresholds.
✅ REAL: The ML techniques
The prediction pipeline uses genuine scikit-learn models (Random Forest + Gradient Boosting ensemble) with proper cross-validation, train/test splits, and AUC-ROC evaluation — the same methods used in published microbiome research.
✅ REAL: The biology
The Learn page content — GABA production by Lactobacillus, butyrate from Faecalibacterium, the vagus nerve pathway, serotonin in the gut — is all established science from the cited research.
⚠️ SIMULATED: The exact numbers in charts
The specific correlation coefficients (e.g., r = −0.38) are not copied from any paper. They emerge from our simulation. The papers say "Coprococcus is significantly depleted in depression" — we coded that as an effect size of −0.40 and generated 500 data points. The resulting chart numbers are plausible but fabricated.
⚠️ SIMULATED: The 500 data points
Every individual sample is computationally generated — no real patients were measured. The dataset is designed so the statistical patterns match research findings, like a flight simulator models real physics without a real airplane.
⚠️ SIMULATED: The effect sizes
We approximated effect sizes from paper descriptions (e.g., "significantly depleted" → −0.40). These are informed estimates, not exact values extracted from the studies. Different approximations would produce different chart numbers.
⚠️ SIMULATED: ML predictions
The model is trained on simulated data, so its predictions reflect those simulated patterns — not clinical validation. It correctly learns which bacteria matter (because we encoded that), but its confidence scores are not clinically validated.
💡 Why this approach still matters
Real patient microbiome data is protected by privacy laws and requires IRB approval to access. By building on published effect sizes rather than raw data, we demonstrate a complete, functional pipeline — from data → visualization → ML prediction → personalized recommendations — that could be plugged into real clinical datasets when partnerships and approvals are in place. The architecture is production-ready; the data is the placeholder.
⚠️ Important Limitations
Transparency is important. Here's what this tool is not:
🚫 Not a Medical Diagnostic Tool
GutMind Explorer is for education and exploration only. Predictions should never replace professional medical advice, diagnosis, or treatment.
🧪 Research Is Evolving
The gut-brain axis is an active area of study. Many associations are correlational, not causal. Effect sizes vary across populations, diets, and methodologies. New papers may update or contradict current findings.
📊 Correlation ≠ Causation
Even in the real studies, lower Coprococcus in depressed patients doesn't prove the bacteria caused better mental health. The relationship could be reversed, or both could be driven by a third factor like diet.
🚀 Where This Could Go
As a prototype, GutMind Explorer demonstrates the potential for accessible microbiome-mental health tools:
Real Data Integration
Partner with microbiome testing companies or research institutions to train on real, anonymized clinical datasets.
API Auto-Import
Connect directly to test kit APIs so users can import their microbiome results with one click instead of manual entry.
Longitudinal Tracking
Let users track their microbiome over time and see how dietary or lifestyle changes affect their gut-brain profile.
Clinical Validation
With proper IRB approval and clinical partnerships, validate the ML model against real patient outcomes.
Built with FastAPI, scikit-learn, and Chart.js
This tool is for educational purposes only. It is not intended to diagnose, treat, cure, or prevent any disease.