FOCUS

FOCUS profiles organisms and estimates their relative abundances from unannotated shotgun metagenomic sequencing reads using a composition-based non-negative least squares (NNLS) approach.

Key Features:

• Scalability: Scales to large metagenomic datasets and accommodates varying read counts and lengths from modern sequencing platforms.
• Non-Negative Least Squares (NNLS) Optimization: Uses NNLS to resolve community composition by minimizing discrepancies between observed and predicted compositional profiles.
• Composition-based profiling: Operates directly on compositional signatures of reads without requiring per-read annotation or alignment to reference genomes.
• Implementation: Implemented in Python.

Scientific Applications:

• Disease Diagnostics: Supports identification of microbial compositions in clinical samples to inform analyses of pathogen presence.
• Microbial Ecology Studies: Enables inference of community structure and dynamics in environmental and host-associated microbiomes.
• Biotechnological Applications: Facilitates monitoring of microbial communities relevant to fermentation, waste degradation, and other bioprocesses.

Methodology:

FOCUS applies a composition-based analysis directly to unannotated shotgun reads and infers organism presence and relative abundances by solving a non-negative least squares (NNLS) optimization problem.