GutSim is an AI-powered digital twin of the human gut. It synthesizes a growing body of published in-vitro research to deliver evidence-based predictions on microbiome composition, SCFAs, pH, gas production, and bloating risk.
Thousands of in-vitro gut model studies exist, but findings are trapped in PDFs with no way to query, synthesize, or predict across them.
Findings scattered across journals. Researchers re-discover what's already known because there's no unified, queryable database of gut model results.
A single in-vitro gut model experiment costs $50K-$150K and takes months. Companies test ingredients that already have published data they can't easily find.
"Will this dose of inulin cause bloating?" has no quick, evidence-based answer. Product teams rely on marketing claims, not integrated evidence.
No system synthesizes evidence across studies to predict how different doses affect different regions of the colon with calibrated confidence.
GutSim continuously ingests published gut model research and builds a structured knowledge base. When you ask a question, AI reasons over the evidence to deliver predictions you can trust.
Enter any ingredient, dose, or combination — "5 g/day inulin", "inulin + LGG", or compare two doses
GutSim searches its knowledge base for relevant measurements, mechanistic rules, and similar studies
Findings are synthesized across multiple studies, weighted by relevance, quality, and dose proximity
Every prediction is assigned a calibrated confidence score based on the strength and breadth of available evidence
Per-compartment predictions for SCFAs, microbiome, pH, gas, bloating risk — with full evidence trail
Every prediction is transparent, dose-aware, and traceable back to the studies that support it.
Knows what doses have been tested. Flags extrapolation beyond studied ranges and adjusts confidence accordingly. Compare multiple doses side-by-side.
"Inulin and LGG" — searches each separately and the combination, with per-ingredient attribution weights showing what drives each prediction.
AI assesses evidence quality, then a deterministic post-processor caps scores by study count and data source type. No over-confident claims.
Every prediction links to the measurements, rules, and studies that support it. See exactly which papers the AI used to reach its conclusion.
Predictions per gut region — ascending, transverse, and descending colon — each with distinct pH, retention time, and microbial profiles.
Bloating risk and fermentation intensity show the formula, input sub-scores, and reasoning. Every number is explainable and auditable.
GutSim's knowledge base is constructed from peer-reviewed in-vitro gut model studies — and growing continuously as new research is published.
GutSim helps anyone who needs to understand how ingredients behave in the human gut before committing to expensive experiments.
Test hypotheses about prebiotics, probiotics, and synbiotics in seconds instead of months. Identify the most promising candidates before committing lab resources.
Use GutSim to identify knowledge gaps, set dose ranges, and validate findings against the existing body of published gut model research.
Generate reports linking product ingredients to published evidence, with confidence scores and full citation trails for regulatory submissions.
Whether you're interested in a demo, a pilot partnership, or research collaboration — we'd love to hear from you.
Thanks for reaching out. We'll get back to you shortly at the email you provided.