Searching match for single-cell open-chromatin profiles in large pools of single-cell transcriptomes and epigenomes for reference supported analysis

Abstract

The true benefits of large data-sets of single-cell epigenome and transcriptome profiles can be availed only when they are searchable to annotate individual unannotated cells. Matching a single-cell epigenome profile to a large pool of reference cells remains as a challenge and largely unexplored. Here, we introduce scEpiSearch, which enables a user to query single-cell open-chromatin read-count matrices for comparison against a large pool of single-cell expression and open-chromatin profiles from human and mouse cells (~ 3.5 million cells). Besides providing accurate search in a short time and scalable visualization of results for multiple query cells, scEpisearch also provides a low-dimensional representation of single-cell open-chromatin profiles. It outperformed many other methods in terms of correct low dimensional embedding of single-cell open-chromatin profiles originating from different platforms and species. Here we show how scEpiSearch is unique in providing several facilities to assist researchers in the analysis of single-cell open-chromatin profiles to infer cellular state, lineage, potency and representative genes.