scShaper: an ensemble method for fast and accurate linear trajectory inference from single-cell RNA-seq data

• DOI: 10.1093/bioinformatics/btab831
• Publication Date: 2021-12-03T20:12:43Z
• Authors: Johannes Smolander; Sini Junttila; Mikko S Venäläinen; Laura L Elo
• Keywords: RNA-Seq

Abstract Motivation Computational models are needed to infer a representation of the cells, i.e. trajectory, from single-cell RNA-sequencing data that model cell differentiation during dynamic process. Although many trajectory inference methods exist, their performance varies greatly depending on dataset and hence there is need establish more accurate, better generalizable methods. Results We introduce scShaper, new method enables accurate linear inference. The ensemble approach scShaper generates continuous smooth pseudotime based set discrete pseudotimes. demonstrate able trajectories for variety trigonometric trajectories, including which commonly used principal curves fails. A comprehensive benchmarking with state-of-the-art revealed achieved superior accuracy ordering and, in particular, differentially expressed genes. Moreover, fast few hyperparameters, making it promising alternative pseudotemporal ordering. Availability implementation available as an R package at https://github.com/elolab/scshaper. test https://doi.org/10.5281/zenodo.5734488. Supplementary information Bioinformatics online.