MCSPACE: inferring microbiome spatiotemporal dynamics from high-throughput co-localization data

Area of Science:

• Microbiome research
• Computational biology
• Systems biology

Background:

• High-throughput sequencing enables large-scale microbiome analysis.
• Microbiome co-localization data present unique challenges due to high dimensionality and noise.
• Understanding gut microbiome spatial structuring is crucial for host-microbial ecosystem function.

Purpose of the Study:

• To develop a probabilistic AI method (MCSPACE) for inferring spatially coherent microbial assemblages from co-localization data.
• To characterize microbial dynamics over time and responses to perturbations.
• To provide an open-source tool for microbiome biogeography research.

Main Methods:

• Development of MCSPACE, a probabilistic AI method for microbiome co-localization data analysis.
• Generation of a large-scale longitudinal microbiome co-localization dataset in mice subjected to dietary perturbations.
• Benchmarking MCSPACE against existing methods using murine and human longitudinal datasets.

Main Results:

• MCSPACE successfully infers spatially coherent microbial assemblages and their dynamics.
• The method demonstrates superior performance compared to existing approaches.
• Identification of temporally persistent and dynamic microbial assemblages in the human gut.
• Revealed shifts in murine gut microbial assemblages induced by specific dietary components.

Conclusions:

• MCSPACE is a powerful tool for elucidating microbiome biogeography and spatiotemporal structuring.
• The method provides insights into the role of spatial relationships in host-microbial ecosystems.
• MCSPACE is available as an open-source software tool for the research community.