Establishing microbial composition measurement standards with reference frames.

Area of Science:

• Microbiome research
• Computational biology
• Statistical analysis

Background:

• Differential abundance analysis in microbiome studies is often controversial.
• Current gold standard methods require laborious total microbial load measurements.
• Most studies rely on relative abundance data, which has known limitations.

Purpose of the Study:

• To demonstrate common pitfalls in comparing relative microbiome abundance across samples.
• To introduce two novel solutions, termed "reference frames", to accurately determine taxonomic shifts.
• To reveal microbial changes without the need to estimate total microbial load.

Main Methods:

• Defining and applying "reference frames" to compositional microbiome data.
• Utilizing an oral time series experiment to test the robustness of reference frames.
• Analyzing two independent published datasets from subjects with atopic dermatitis.

Main Results:

• Reference frames alleviate false positives and yield consistent results on both raw and cell-count normalized data.
• Reference frames identified consistent, differentially abundant microbes previously undetected in atopic dermatitis datasets.
• The proposed methods allow reassessment of published relative abundance data.

Conclusions:

• Reference frames offer a powerful and intuitive approach to analyzing compositional microbiome data.
• These methods overcome common pitfalls associated with relative abundance analysis.
• The approach enables the revelation of reproducible microbial changes from standard sequencing data without new assays.