metagenomeSeq

metagenomeSeq identifies differentially abundant features, such as Operational Taxonomic Units (OTUs) or species, across microbial marker-gene surveys by normalizing for sequencing depth and modeling undersampling in sparse high-throughput datasets.

Key Features:

• Normalization Technique: Employs a normalization method tailored to sparse marker-gene datasets that adjusts for sequencing depth and other technical biases.
• Statistical Modeling: Implements a statistical model that accounts for undersampling effects to improve differential abundance inference from sparse data.
• Performance: Demonstrated superior performance on simulated data and published microbiota datasets for detecting disease associations and feature correlations.

Scientific Applications:

• Differential-abundance analysis: Identifying differentially abundant OTUs or species in studies of microbial community composition.
• Microbiome–disease associations: Detecting disease associations and feature correlations in contexts such as gastrointestinal disorders, metabolic diseases, and immune responses.
• Large-scale marker-gene surveys: Analyzing large-scale marker-gene survey datasets characterized by sparse counts and variable sequencing depth.

Methodology:

Data are first normalized using the tool's normalization technique to adjust for sequencing depth and other technical factors. A statistical model tailored to account for undersampling effects is then applied to perform differential abundance analysis.