Evaluating the impact of sequencing depth on transcriptome profiling in human adipose

Area of Science:

• Genomics and Transcriptomics
• Molecular Biology
• Bioinformatics

Background:

• RNA sequencing (RNA-Seq) offers advanced transcriptomic variation detection beyond microarrays.
• Understanding transcriptomic changes in cardiometabolic diseases under stress and therapy is critical.
• Optimal RNA-Seq experimental design necessitates knowledge of sequencing depth's impact on results.

Purpose of the Study:

• To investigate the relationship between RNA sequencing depth and the detection of gene expression and alternative splicing variations.
• To determine the required sequencing depth for identifying differential expression (DE) and differential alternative splicing (DAS) in response to endotoxin (LPS) in human adipose tissue.

Main Methods:

• Utilized deeply sequenced Illumina HiSeq 2000 101 bp paired-end RNA-Seq data from human adipose tissue.
• Analyzed samples from a healthy individual before and after systemic lipopolysaccharide (LPS) administration.
• Evaluated the impact of varying sequencing depths on the detection of expressed genes, alternative splicing events, DE, and DAS.

Main Results:

• Approximately 100-150 million filtered reads are needed to detect expressed genes and alternative splicing events.
• Detecting 80% of differential expression (DE) events requires around 300 million filtered reads.
• Detecting 80% of differential alternative splicing (DAS) events necessitates at least 400 million filtered reads, indicating a higher requirement than for DE.

Conclusions:

• While modest sequencing depths (∼100 M reads) detect most genes and AS events, they yield less accurate expression level estimations.
• Reliable detection of differential alternative splicing (DAS) in human adipose tissue requires significantly higher sequencing depth than differential gene expression (DE).
• This study provides crucial insights for designing RNA-Seq experiments and interpreting transcriptomic data in cardiometabolic disease research.