Stochastic Analysis of the RT-PCR Process in Single-Cell RNA-Seq

Abstract

The single-cell RNA-seq allows exploring the transcriptome for one cell at a time. By doing so, cellular regulation is pictured. One limitation is the dropout events phenomenon, where a gene is observed at a low or moderate expression level in one cell but not detected in another. Dropouts obscure legitimate biological heterogeneity leading to the description of a small fraction of the meaningful relations. We used a stochastic approach to model the Reverse Transcription Polymerase Chain Reaction (RT-PCR) kinetic, in which we contemplated the temperature profile, RT-PCR duration, and reaction rates. By studying the underlying biochemical processes of RT-PCR, using a computational and analytical framework, we show a minimal amount of RNA to avoid dropout events. We further use this fact to characterize the limits in the dispersion reduction. Dispersion asymptotically decreases as the RNA initial value increases. Despite always being a basal dispersion, their decreasing speed is modulated mainly by the degradation rates, particularly for the RNA. We concluded that the critical step into the RT-PCR is the RT phase due to the fragile nature of the RNA. We propose that limiting RNA degradation might ensure that the portraited transcriptional landscape is unbiased by technical error.