SIAMCAT

SIAMCAT performs association analysis of microbial community profiles with host phenotypes using machine-learning and statistical methods on metagenomic sequencing data.

Key Features:

• R implementation: Implemented in R for computational analysis of microbiome data.
• Machine Learning Integration: Provides workflows to train and evaluate classifiers for biomarker extraction and validation.
• Statistical Testing Workflows: Includes statistical testing workflows with customizable options for robust inference.
• Modular Design: Modular architecture enabling tailored analytical workflows.
• Metagenomic Sequencing Analysis: Processes metagenomic sequencing data from fecal samples to identify taxonomic markers.
• Validation Across Populations: Validates findings across independent patient and control cohorts from different countries.
• Metabolic Shift Analysis: Infers metabolic shifts in microbial communities, including transitions from fiber degradation to host carbohydrate and amino acid utilization and increased lipopolysaccharide metabolism in CRC.

Scientific Applications:

• Taxonomic marker discovery: Identify taxonomic markers that distinguish disease states from healthy controls.
• Colorectal carcinoma studies: Analyze fecal metagenomic sequencing to detect CRC-associated microbiota changes and markers.
• Biomarker-based screening: Support biomarker discovery and validation for cancer screening, with reported accuracy comparable to fecal occult blood tests (FOBT) and increased sensitivity when combined.
• Microbial metabolic insights: Characterize disease-associated metabolic shifts in microbial communities, such as increased lipopolysaccharide metabolism in CRC.
• General microbiota-host research: Apply to other diseases where microbiota-host interactions are relevant.

Methodology:

Machine-learning workflows to train and evaluate classifiers, statistical testing workflows, processing of metagenomic sequencing data to identify taxonomic markers, validation across independent cohorts, and inference of metabolic shifts.