zFPKM

The zFPKM metric, which is a normalization tool for RNA-seq experiments, is introduced in this study. It offers a refined method for distinguishing between biologically active gene expressions and those likely arising from biological or experimental noise. The inception of RNA-seq technology revealed a vast mammalian transcriptome, sparking initial enthusiasm about the potential to measure transcripts from nearly all known genes through ultradeep sequencing. However, skepticism grew regarding the biological significance of low-abundance transcripts, with concerns that these might largely reflect noise rather than functional gene expression. This led to the adoption of arbitrary expression thresholds (e.g., 0.3 - 1 FPKM) for distinguishing meaningful gene expression in downstream analysis.

Advancements in library preparation, sequencing technologies, and analytical methods have significantly mitigated earlier concerns about RNA-seq reliability, especially for low-abundance transcripts. Leveraging genomic data from extensive studies like ENCODE, which offers orthogonal information through chromatin state data, the study behind zFPKM provides a modern perspective on RNA-seq's capability to discern active from inactive gene expressions accurately.

This normalization method has significant implications for RNA-seq study design, suggesting that a read depth of twenty to thirty million mapped reads is sufficient for high-confidence quantification of genes expressed above the noise threshold. By using ENCODE chromatin state data for validation, zFPKM enhances the accuracy of distinguishing biologically relevant gene expressions and sets a new standard for evaluating RNA-seq data.