Learning functional groups in complex microbiomes.

Area of Science:

• Microbiology
• Computational Biology
• Systems Biology

Background:

• Microbial communities in soil and gut perform vital functions.
• Understanding the link between microbial composition and function is challenging due to complexity.

Purpose of the Study:

• To develop a data-driven approach for simplifying microbial communities.
• To identify key microbial groups and genes driving community functions.
• To create structure-function maps for microbiomes.

Main Methods:

• Utilized a neural-network based clustering algorithm.
• Applied the method to gut metagenomes, ocean metagenomes, and soil bacterial species.
• Integrated interpretable machine learning (ML) with strain isolation and sequencing.

Main Results:

• Successfully recovered known functional groups in gut communities.
• Distilled ~500 gene modules into three sparse groups in ocean metagenomes.
• Reduced ~4400 soil bacterial species into two groups for nitrate metabolism modeling.
• Connected microbial group specialization to community responses to perturbations.

Conclusions:

• The approach simplifies complex microbiomes into interpretable structure-function maps.
• Enables discovery of molecular mechanisms underlying health and environmental processes.
• Demonstrates a method for function-informed dimensionality reduction in biological systems.