Technology

The intelligence layer for antimicrobial decision-making

Biologically Explainable AI. Clinically Actionable Insights.

We map whole-genome mutational signatures to forecast resistance-evolution risk—giving clinicians the foresight to help prevent treatment failures, not just react to them.

Platform

How It Works

A longitudinal learning system that links bacterial evolution to patient outcomes

01

Forecast

Forecast treatment-emergent resistance risk and detect current antimicrobial susceptibility profile

02

Act

Actionable insights within 24 hours—designed to integrate into antimicrobial stewardship workflows for timely decisions to optimize therapy.

03

Learn

Turn every case into a feedback loop. Link pathogen evolution signals with the antimicrobial course and downstream clinical outcomes to generate new decision signals.

Science

The Science Behind Our Platform

Pioneering bacterial mutational signature mapping for resistance forecasting

novel biomarkers

Genome-wide evolution signals

We pioneered the application of mutational signature analysis to antimicrobial resistance. By analyzing whole-genome patterns rather than individual genes, we detect:

  • Resistance patterns independent of specific known mutations
  • Novel and emerging resistance mechanisms
  • Evolutionary trajectories toward resistance acquisition
AI

Mechanism-linked AI

Our AI isn't a black box. Trained on massive longitudinal datasets linking mutational signatures to resistance phenotypes and clinical outcomes from real patients, every forecast is grounded in biology:

  • Predictions tied to specific evolutionary biological mechanisms
  • Transparent rationale for actionable insights
  • Continuous learning from real-world patient outcomes
Enabled by

A New Class of Predictive Genomic Biomarkers

Our AI is enabled by mutational signatures—fundamentally new biomarkers that act as "fingerprints" of errors left behind in DNA by cellular processes.

Unlike conventional methods that screen for specific known mutations, our approach detects resistance independent of specific, previously identified genetic changes.

Key Advantage

This mechanism-independent approach enables detection of novel and emerging resistance patterns that traditional diagnostics would miss entirely.

01

Genome Sequencing

Extract and sequence bacterial DNA from patient sample

02

Signature Analysis

ML models identify resistance-associated mutational patterns

03

Resistance Profile

Generate comprehensive drug susceptibility report

04

Clinical Decision

Clinician selects optimal antimicrobial for patient

Validation

Evidence & Recognition

Peer-reviewed research and industry recognition supporting our technology

Nature Communications
Published Research

Nature Communications

Mutational signature analysis predicts bacterial hypermutation and multidrug resistance

IMARI Conference
Upcoming

IMARI 2026

Presenting our latest research at the Interdisciplinary Meeting on Antimicrobial Resistance & Infection

ID Week
Conference

ID Week 2023

Presented findings at the leading infectious disease conference

ASM Microbe
Conference

ASM Microbe 2023

Presented at the American Society for Microbiology annual meeting

National Science Foundation
Funding

NSF Funded

Backed by the National Science Foundation for breakthrough innovation