Original title: DeconV: Probabilistic Cell Type Deconvolution from Bulk RNA-sequencing Data
Authors: Artur Gynter,Dimitri Meistermann,Harri Lahdesmaki,Helena Kilpinen
In this article, the authors discuss the challenges associated with accurately estimating the proportions of different cell types in bulk RNA-Seq samples. While bulk RNA-Seq is a commonly used method for gene expression profiling, it does not offer single-cell resolution. To address this issue, the authors introduce a new framework called DeconV. This framework uses single-cell RNA-Seq data as a reference to improve cell-type deconvolution in bulk samples. DeconV incorporates statistical frameworks developed for single-cell RNA-Seq, simplifying issues related to reference preprocessing. The authors compared DeconV to other established methods and found that it performs similarly in terms of accuracy. However, DeconV provides additional interpretability by offering confidence intervals for its predictions. The modular design of DeconV also allows for the investigation of discrepancies between bulk-sequenced samples and artificially generated pseudo-bulk samples. Overall, DeconV offers a promising solution for accurately estimating cell-type proportions in bulk RNA-Seq data.
Original article: https://www.biorxiv.org/content/10.1101/2023.12.07.570524v1