PREMISE: A Quality-Aware Probabilistic Framework for High-Resolution Source Assignment in Metagenomic Sequencing

PREMISE achieves the highest species-level read classification F1 score (0.992) on ten dense CAMI II strain-madness metagenomic samples, with no false-positive or missed species.

• Why it matters: Existing k-mer approaches fragment reads and ignore per-base quality and intra-read linkage, which confounds closely related variants and limits real-time pathogen surveillance.
• What they did: The Rust framework aligns whole reads using a bidirectional run-length FM-index and performs a quality-aware expectation-maximization abundance estimation mapped over a taxonomy, with an optional sparsity penalty.
• The result: At strain level, it attains top precision (0.80 vs at most 0.64 for competitors), achieves 100% recall, and estimates abundances as accurately as the best competing tool.

AI summary of the abstract · may contain errors · not medical advice · read the paper.