Gene Trajectory Inference for Single-cell Data by Optimal Transport Metrics

Autor: Rihao Qu, Xiuyuan Cheng, Esen Sefik, Sarah Platt, James Garritano, Ian Odell, Richard A. Flavell, Peggy Myung, Yuval Kluger
Rok vydání: 2022
Popis: Single-cell RNA-sequencing has enabled researchers to study dynamic biological processes by inferring cell trajectories. Current cell trajectory methods attempt to utilize whole transcriptomic profiles to draw connections among cells. However, whole-transcriptome data incorporates multiple gene programs, some of which are nonspecific or irrelevant to the process of interest. This results in fuzzy cell trajectories that fail to reveal the true progression of the process studied. Here, we present GeneTrajectory, a new approach that deconvolves complex biological processes by identifying trajectories of genes rather than trajectories of cells. Specifically, the distances between gene expression distributions over the cell graph are computed using an optimal-transport metric to construct gene trajectories. GeneTrajectory unbiasedly extracts gene programs and their gene pseudotemporal order from whole-transcriptome data without relying on cell trajectory inferences. Using published and novel datasets, we demonstrate that GeneTrajectory identifies the gene dynamics of biological processes that cannot be effectively resolved by cell trajectory approaches.
Databáze: OpenAIRE