Researchers led by Professor Jingyi Li from the Statistics Department at UCLA have developed a simulator that allows researchers to optimally design single-cell RNA sequencing experiments in a cost-effective manner.
Single-cell RNA sequencing (scRNA-seq) is a promising technology used for biomarker discovery, drug development, and precision medicine in both academia and industry. ScRNA-seq allows researchers to identify gene expression in individual cells, as opposed to capturing an average gene expression profile in a mixture of cells. This has the potential to identify cell subtypes and reveal signature genes that characterize them. It has been used to understand stem cell differentiation, embryogenesis, neurological disorders, and tumorigenesis. However, effective design of scRNA-seq experiments remains a theoretical and computational challenge, and performing pre-experiments to assist the design can be very costly. Often times, researchers have to balance between using fewer cells to get highly accurate data and using a greater number of cells for a broader survey of gene expression. Currently, there are software packages that can simulate data to assist experimental design, but they cannot evaluate or compare different scRNA-seq experimental protocols, nor can they rationally optimize a design once a protocol has been chosen.
Researchers led by Professor Jingyi Li from the Statistics Department at UCLA have developed a simulator, scDesign, that allows researchers to optimally design single-cell RNA sequencing experiments in a cost-effective manner. Their innovation is able to generate synthetic scRNA-seq datasets and use them to assist the scRNA-seq experimental design. ScDesign generates more accurate synthetic data than its competitors along with answering key experimental design questions such as how many cells need to be used to maximize the power (or other accuracy criteria) in detecting differentially expressed genes between two cell groups. Furthermore, scDesign gives researchers the flexibility of comparing and contrasting different experimental protocols and different machines used for scRNA-seq that are currently on the market. The figure below shows how scDesign can help design scRNA-seq experiments.
genomics, transcriptomics, software, RNA-seq, scRNA-seq, single cell, RNA, bioinformatics, experimental design, simulation