MDSINE

MDSINE infers dynamical systems models from microbiome time-series data to predict temporal behaviors of host-associated microbial communities and inform rational design of bacteriotherapies.

Key Features:

• Algorithmic Suite: A suite of algorithms tailored to analyze complex microbial dynamics over time.
• Performance Superiority: Demonstrated via simulations to outperform existing inference methods in accuracy and robustness.
• Forecasting Capabilities: Capable of accurately forecasting microbial community dynamics under varying conditions.

Scientific Applications:

• Gnotobiotic Mice Studies: Applied to gnotobiotic mice datasets including analyses of Clostridium difficile infections and responses to immune-modulatory probiotics.
• Pathogen Growth Inhibition: Identifies stable sub-communities within the microbiome that can inhibit pathogen growth.
• Community Integrity Analysis: Pinpoints bacteria crucial for maintaining community integrity under perturbations.

Methodology:

MDSINE constructs dynamical interaction models from microbiome time-series data and uses simulations to validate predictive accuracy across experimental setups such as infections and probiotic interventions.