MARS-seq2.0 an experimental and analytical pipeline for indexed sorting combined with Single-cell RNA sequencing

Human tissues are composed of trillions of cells that populate a complex space of molecular phenotypes and functions that vary in abundance by 4-9 orders of magnitude. Relying solely on unbiased sampling to characterize cellular niches becomes non-feasible, as the marginal utility of collecting more cells is diminishing quickly. Further, in many clinical samples, the relevant cell types are scarce and efficient processing is critical. We developed an integrated pipeline for index-sorting and massively parallel single-cell RNA-seq (MARS-seq2.0) that is based on over 1 million cells sequenced with this pipeline. These include, identification of unique cell types across different tissues, diseases, and unique model systems and organisms. In addition we developed several modalities that are based on MARS-seq2.0, including genetic perturbations (CRISP-seq) and spatial reconstruction (NICHE-seq). Here we present a detailed step-by-step procedure to apply the method. We combine sub microliter reaction volumes, optimization of enzymatic mixtures and enhanced analytical pipeline (http://compgenomics.weizmann.ac.il/tanay/?page_id=672) to significantly lower cost, improve reproducibility and reduce the well-to-well contamination. Data analysis combines multiple layers of quality assessment and error detection and correction, graphically presenting key statistics of library complexity, noise distribution and sequencing saturation. Importantly, our combined FACS and single cell RNA-seq workflow enables for intuitive approaches to deplete or enrich for cell populations in a data driven manner that is essential for efficient sampling of complex tissues. The experimental protocol, from sorted plates to a ready to sequence library, takes 2 days. Sequencing and running the analytical pipeline takes another 1-2 days. Overall design: Transcriptional profiling of single cells from cell lines and mouse models of, generated from deep sequencing of thousands of cells, sequenced in several batches on illumina Nextseq500. For the BM experiments (samples 1-16), cells were extracted from the BM of 4 individual mice, enriched for the eindicated population and sorted for MARS-seq2.0 cell capture plates. During library preparation, in order to calibrate and optimize the MARS-seq protocol, we have used different compositions of secon strand synthesis mixes as described in the 'description' column. For the mixed species experiments (samples 17-20), human and mouse cell lines were disocciated and mixed in 1:1 ratio. For the cell line experiments (samples 21-30), either ES, MEF or a mixture of ES and MEF cell lines, in at least duplicates, were disocciated and sorted into MARS-seq2.0 cell capture plates. Different conditions were used in the library prepartion protocol in order to optmize the protocol, as described in the 'description' column. The 'metadata.txt' file associating each single cell with its amplification batch and index sorting readouts is provided.