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

Area of Science:

• Computational biology
• Single-cell genomics
• Bioinformatics

Background:

• Accurate trajectory inference from single-cell RNA-sequencing (scRNA-seq) data is crucial for modeling dynamic biological processes like cell differentiation.
• Existing trajectory inference methods exhibit variable performance across different datasets, highlighting the need for more robust and generalizable approaches.

Purpose of the Study:

• To introduce scShaper, a novel computational method for accurate linear trajectory inference from scRNA-seq data.
• To evaluate scShaper's performance against state-of-the-art methods, particularly for modeling cell differentiation.

Main Methods:

• scShaper employs an ensemble approach to generate a continuous pseudotime from discrete pseudotimes, facilitating smooth trajectory representation.
• The method was tested on various trigonometric trajectories and benchmarked against established trajectory inference techniques.

Main Results:

• scShaper accurately infers linear trajectories, outperforming the principal curves method on challenging datasets.
• Benchmarking demonstrated scShaper's superior accuracy in cell ordering and identification of differentially expressed genes.
• scShaper is computationally efficient and requires minimal hyperparameter tuning.

Conclusions:

• scShaper provides a highly accurate and efficient method for linear trajectory inference in scRNA-seq data.
• Its robust performance makes it a valuable alternative to existing methods for pseudotemporal ordering and analysis of cell differentiation.