SUMMARY: UCLA researchers in the Department of Statistics have developed a software to compare existing doublet detection algorithms for Single-Cell RNA Sequencing Data.
BACKGROUND: Single-cell RNA sequencing (scRNA-seq) provides the expression profiles of individual cells and is considered the gold standard for defining cell states and phenotypes as of 2020. scRNA-Seq is becoming widely used across biological disciplines including Developmental biology, Neurology, Oncology, Immunology, Cardiovascular research and Infectious disease.This technology encompasses a group of methods that provide transcriptional information for individual cells. By profiling individual cells, this technology allows researchers to identify the intricate molecular differences that make cells unique. scRNA-seq works by taking a sample of cells and dissociating them to individual cells. During this process it is common for not all cells to properly dissociate. This results in what is known as a doublet or generating transcriptional information for two cells at once rather than one. Since scRNAseq is a data driven technology, the presence of these doublets can lead to incorrect conclusions. Numerous methods have been created to identify and even correct the presence of doublets within scRNAseq data. Currently there are no comprehensive tools to allow the fair comparison of these methods. Especially as the technology breaks into the medical field, there is a present need to identify the best methods for doublet-detection.
INNOVATION: Researchers at UCLA led by Dr. Jingyi Jessica Li have developed novel software to compare existing doublet detection algorithms. It shows that existing methods exhibited diverse performance and distinct advantages in different aspects. Currently, there are no alternative software options.
POTENTIAL APPLICATIONS:
ADVANTAGES:
DEVELOPMENT-TO-DATE: This software has been validated by testing multiple cutting-edge doublet detection methods using both real and synthetic datasets.
RELATED PAPERS