Misuse of RPKM or TPM normalization when comparing across samples and sequencing protocols

Area of Science:

• Genomics
• Bioinformatics
• Molecular Biology

Background:

• RNA-sequencing (RNA-seq) is a key technology for transcriptome profiling.
• Gene expression levels are quantified using metrics like RPKM (reads per kilobase of transcript per million reads mapped) and TPM (transcripts per million).
• These metrics account for gene length and sequencing depth.

Purpose of the Study:

• To clarify common misconceptions regarding the comparability of RPKM and TPM values across different RNA-seq experiments.
• To highlight the limitations of RPKM and TPM when comparing gene expression across samples with varying RNA compositions or sequencing protocols.
• To raise awareness among scientists about the potential misuse of these normalization metrics.

Main Methods:

• Review of existing literature and common practices in RNA-seq data analysis.
• Illustration of scenarios where RPKM and TPM values can lead to incorrect interpretations.
• Discussion of the impact of sample RNA composition and sequencing protocols on normalization metrics.

Main Results:

• RPKM and TPM measure relative transcript abundance within a specific sample's sequenced population.
• These values are not inherently normalized for comparison across samples with different RNA compositions or experimental conditions.
• Differences in experimental conditions or sequencing protocols can significantly alter the sequenced RNA repertoire, affecting comparability.

Conclusions:

• Scientists should exercise caution when comparing RPKM and TPM values across different RNA-seq projects or protocols.
• Understanding the dependency of RPKM and TPM on sample RNA composition is crucial for accurate data interpretation.
• Further investigation into robust cross-sample normalization methods for RNA-seq data is warranted.