Identification of new marker genes from plant single-cell RNA-seq data using interpretable machine learning methods

Abstract

An essential step in the analysis of single-cell RNA sequencing data is to classify specific cell types with marker genes. In this study, we have developed a machine learning pipeline called Single cell Predictive markers (SPmarker) to identify novel cell-type marker genes in the Arabidopsis root. Unlike traditional approaches, our method uses interpretable machine learning methods to select marker genes. We have demonstrated that our method can (1) assign cell types based on cells that were labeled using published methods, (2) project cell types identified by trajectory analysis from one dataset to other datasets, and (3) assign cell types based on internal GFP markers. Using SPmarker, we have identified hundreds of new marker genes that were not identified before. As compared to known marker genes, we have found more orthologous genes of these new marker genes in corresponding rice single cell clusters. We have also found 172 new marker genes for trichoblast in five non-Arabidopsis species, which expands number of marker genes for this cell type by 35-154%. Our results represent a new approach to identify cell-type marker genes from scRNA-seq data and pave the way for cross-species mapping of scRNA-seq data in plants.