Comparison, calibration, and benchmarking of high-throughput single cell RNA-Seq techniques for unbiased cell-type classification

The Human Cell Atlas project seeks to identify and functionally characterize all cell types in the human body. Single-cell RNA-sequencing (scRNA-seq) methods quantify gene expression in individual cells. In principle, highly parallelized scRNA-seq approaches enable marker-free decomposition of cells from complex tissues to identify known and unknown cell types and provide direct insights into regulatory and functional states of each cell. Different scRNA-seq technologies vary in cell capture efficiency, library preparation, and throughput. Importantly, these approaches also vary in sensitivity and accuracy of mRNA quantification. The Human Cell Atlas project will collate data generated across many labs, necessitating the evaluation of experimental replicability, and comparing different experimental procedures. It is important to develop robust sequencing protocols and benchmark their performance on select cells and tissues. This project will perform a systematic comparison of three single cell RNA sequencing technologies, viz., Drop-Seq, Fluidigm C1, and DroNC-Seq and resulting datasets. Droplet-based methods facilitate high-throughput processing of tens of thousands of cells in parallel and at low costs, making them ideal for the Human Cell Atlas project. This project proposes to compare the sensitivity, accuracy, and precision with which the three methods quantify transcript levels within individual cells. In addition, this project will directly compare these methods using a ‘synthetic tissue’ created from a mixture of multiple cell types at known ratios.