Unifying the analysis of high-throughput sequencing datasets: characterizing RNA-seq, 16S rRNA gene sequencing and...

Area of Science:

• Bioinformatics
• Computational Biology
• Statistical Genetics

Background:

• High-throughput sequencing generates complex datasets (e.g., RNA-seq, ChIP-seq, 16S rRNA gene sequencing) with read counts mapped to numerous features.
• Current data analysis methods are experiment-specific and lack cross-applicability.
• Compositional data analysis (CDA) methods, treating data as relative abundances, offer more robust and reproducible analyses.

Purpose of the Study:

• To evaluate the applicability and robustness of the ALDEx2 tool for analyzing diverse high-throughput sequencing data.
• To demonstrate the effectiveness of CDA in unifying analysis across different sequencing experimental designs.

Main Methods:

• Utilized ALDEx2, a Bayesian-based compositional data analysis tool.
• Applied ALDEx2 to three distinct datasets: in vitro selective growth experiment, RNA-seq experiment, and Human Microbiome Project 16S rRNA gene abundance data.
• Assessed ALDEx2's ability to identify differential features, differentially expressed genes, and distinguishing taxa.

Main Results:

• ALDEx2 accurately identified differential abundance in the selective growth experiment.
• It identified a comparable set of differentially expressed genes in RNA-seq data as existing leading tools.
• ALDEx2 successfully distinguished taxa differentiating tongue dorsum and buccal mucosa in the Human Microbiome Project dataset.
• The tool's design minimizes false positives in datasets with many features and few samples.

Conclusions:

• The ALDEx2 R package is a versatile and robust tool for statistical analysis of high-throughput sequencing count data.
• It is suitable for RNA-seq, 16S rRNA gene sequencing, and differential growth experiments, with potential for other similar techniques.
• Compositional data analysis provides a unified and more reliable approach for diverse sequencing data types.