RandomReadsMG: Rapid, realistic metagenome simulation at terabase scale for benchmarking and experimental design.

RandomReadsMG generates terabase-scale metagenomic datasets within 6 hours, supporting realistic benchmarking and experimental design at large scale.

• Why it matters: Synthetic metagenomes are crucial for testing bioinformatics tools, guiding experiments, and training AI models, but existing methods struggle with speed and realism at scale.
• What they did: The open-source simulator creates communities of up to 50,000 genomes with customizable profiles, sequencing error models, and library artifacts, supporting Illumina, ONT, and PacBio presets.
• The result: It enables rapid, reproducible large-scale data generation and accurate simulation of genome coverage, aiding in threshold estimation for pathogen detection and MAG recovery.