Single-cell RNA sequencing algorithms underestimate changes in transcriptional noise compared to single-molecule RNA...

Area of Science:

• Molecular Biology
• Genomics
• Cell Biology

Background:

• Stochastic fluctuations, or noise, in gene transcription are a major driver of cell-to-cell variability.
• Accurate quantification of genome-wide transcriptional noise is crucial but remains challenging.
• Understanding noise is essential for deciphering cellular heterogeneity and function.

Purpose of the Study:

• To evaluate and compare various single-cell RNA sequencing (scRNA-seq) algorithms for quantifying transcriptional noise.
• To assess the efficacy of 5'-iodo-2'-deoxyuridine (IdU) as a tool to amplify transcriptional noise for analysis.
• To validate scRNA-seq noise quantification with single-molecule RNA fluorescence in situ hybridization (smFISH).

Main Methods:

• Utilized 5'-iodo-2'-deoxyuridine (IdU) to amplify transcriptional noise in human and mouse cells.
• Applied multiple scRNA-seq data analysis algorithms to quantify genome-wide noise.
• Compared scRNA-seq noise quantification with results from single-molecule RNA fluorescence in situ hybridization (smFISH) for selected genes.

Main Results:

• Most tested scRNA-seq algorithms reported amplified noise for approximately 90% of genes without altering mean expression levels.
• smFISH analysis confirmed noise amplification for the majority of genes examined.
• All scRNA-seq algorithms tended to underestimate noise changes compared to smFISH.

Conclusions:

• Most scRNA-seq algorithms are suitable for quantifying transcriptional noise, even with simple normalization.
• IdU-induced noise enhancement is a globally penetrant phenomenon, increasing noise homeostatically.
• This IdU-based approach offers a promising method for investigating the physiological impacts of transcriptional noise.