GRASE

GRASE estimates the sequencing effort required to recover metagenome-assembled genomes (MAGs) from microbial communities by modeling community structure (richness, evenness, genome size) and combining an analytical coupon collector model with simulations and read subsampling.

Key Features:

• Analytical Model: GRASE employs a coupon collector equation to predict sequencing effort as a function of community richness, evenness, and genome size.
• Simulation-Based Insights: Simulations demonstrate that community properties influence total sequencing effort while the effort to recover an individual genome is primarily determined by that genome's relative abundance and genome size.
• Data-Driven Analysis: GRASE subsamples preexisting sequencing read datasets across varying sequencing efforts and completeness levels to evaluate MAG quantity, completeness, and contamination.
• Diminishing Returns Insight: The approach quantifies diminishing returns in MAG binning success for typical sequencing efforts (1 to 10 Gbp) beyond specific thresholds.
• Sequencing Planning: GRASE enables planning of sequencing experiments based on target genome relative abundance rather than arbitrary sequencing depth to align recovery goals with metabolic potential characterization.

Scientific Applications:

• Sequencing experiment design: Inform sequencing depth decisions to optimize MAG recovery and allocate resources efficiently.
• Targeted recovery of rare genomes: Prioritize sequencing to recover low-abundance genomes using modeled relative abundance and genome size relationships.
• Functional/metabolic potential characterization: Estimate effort required to recover genomes needed for metabolic and functional analyses of microbial communities.
• Environmental and applied microbiome studies: Plan metagenomic sequencing for investigations in health, agriculture, and Earth system processes.

Methodology:

Uses an analytical coupon collector equation, simulation approaches varying community richness, evenness, and genome size, subsampling of existing sequence read datasets to multiple completeness levels, and evaluation of MAG quantity, completeness, and contamination to relate sequencing effort to relative abundance and genome size.