Characterizing the transcriptome of individual cells is fundamental to understanding complex biological systems. We describe a droplet-based system that enables 3′ mRNA counting of tens of thousands of single cells per sample. Cell encapsulation, of up to 8 samples at a time, takes place in ∼6 min, with ∼50% cell capture efficiency. To demonstrate the system's technical performance, we collected transcriptome data from ∼250k single cells across 29 samples. We validated the sensitivity of the system and its ability to detect rare populations using cell lines and synthetic RNAs. We profiled 68k peripheral blood mononuclear cells to demonstrate the system's ability to characterize large immune populations. Finally, we used sequence variation in the transcriptome data to determine host and donor chimerism at single-cell resolution from bone marrow mononuclear cells isolated from transplant patients.

40Characterizing the transcriptome of individual cells is fundamental to understanding complex 41 biological systems. We describe a droplet-based system that enables 3' mRNA counting of up 56peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/065912 doi: bioRxiv preprint first posted online 84 RESULTS 86Droplet-based platform enables barcoding of tens of thousands of cells 88The scRNA-seq microfluidics platform builds on the GemCode ® technology, which has 89 been used for genome haplotyping, structural variant analysis and de novo assembly of a 90human genome [10][11][12] . The core of the technology is a Gel bead in Emulsion (GEM). GEM 91 peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/065912 doi: bioRxiv preprint first posted online 5 generation takes place in an 8-channel microfluidic chip that encapsulates single gel beads at ~80% fill rate (Fig. 1a-c). Each gel bead is functionalized with barcoded oligonucleotides that 93 consist of: i) sequencing adapters and primers, ii) a 14bp barcode drawn from ~750,000 94 designed sequences to index GEMs, iii) a 10bp randomer to index molecules (unique molecular 95 identifier, UMI), and iv) an anchored 30bp oligo-dT to prime poly-adenylated RNA transcripts 96 (Fig. 1d). Within each microfluidic channel, ~100,000 GEMs are formed per ~6-min run, 97encapsulating thousands of cells in GEMs. Cells are loaded at a limiting dilution to minimize co- 98occurrence of multiple cells in the same GEM. 100Cell lysis begins immediately after encapsulation. Gel beads automatically dissolve to 101 release their oligonucleotides for reverse transcription of poly-adenylated RNAs. Each cDNA 102 molecule contains a UMI and shared barcode per GEM, and ends with a template switching 103 oligo at the 3' end (Fig. 1e). Next, the emulsion is broken and barcoded cDNA is pooled for 104PCR amplification, using primers complementary to the switch oligos and sequencing adapters. Methods, Fig. 1f). Briefly, 98-nt of Read1s were aligned against the union of human (hg19) 123peer-reviewed) is the author/funder. All rights reserved. No reuse allowed without permission.The copyright holder for this preprint (which was not . http://dx.doi.org/10.1101/065912 doi: bioRxiv preprint first posted online 6 Based on the distribution of total UMI counts for each barcode (Online Methods), we 124 estimated that 1,012 GEMs contained cells, of which 482 and 538 contained reads that mapped 125 primarily to the human and mouse transcriptome, respectively (and will be referred to as human 126 and mouse GEMs) (Fig. 2a). >83% of UMI counts were associated with cell barcodes, 127indicating low background of cell-free RNA. Eight cell-containing GEMs had a substantial 128 fraction of human and mouse UMI counts (the UMI count is >1% of each species' UMI...

Cancer progression is driven by the accumulation of a small number of genetic alterations. However, these few driver alterations reside in a cancer genome alongside tens of thousands of additional mutations termed passengers. Passengers are widely believed to have no role in cancer, yet many passengers fall within protein-coding genes and other functional elements that can have potentially deleterious effects on cancer cells. Here we investigate the potential of moderately deleterious passengers to accumulate and alter the course of neoplastic progression. Our approach combines evolutionary simulations of cancer progression with an analysis of cancer sequencing data. From simulations, we find that passengers accumulate and largely evade natural selection during progression. Although individually weak, the collective burden of passengers alters the course of progression, leading to several oncological phenomena that are hard to explain with a traditional driver-centric view. We then tested the predictions of our model using cancer genomics data and confirmed that many passengers are likely damaging and have largely evaded negative selection. Finally, we use our model to explore cancer treatments that exploit the load of passengers by either (i) increasing the mutation rate or (ii) exacerbating their deleterious effects. Though both approaches lead to cancer regression, the latter is a more effective therapy. Our results suggest a unique framework for understanding cancer progression as a balance of driver and passenger mutations.

Cancer growth is a multi-stage, stochastic evolutionary process. While cancer genome sequencing has been instrumental in identifying the genomic alterations that occur in human tumors, the consequences of these alterations on tumor growth remains largely unexplored. Conventional genetically engineered mouse models enable the study of tumor growth in vivo, but they are neither readily scalable nor sufficiently quantitative to unravel the magnitude and mode of action of many tumor suppressor genes. Here, we present a method that integrates tumor barcoding with ultra-deep barcode sequencing (Tuba-seq) to interrogate tumor suppressor function in mouse models of human cancer. Tuba-seq uncovers genotype-dependent distributions of tumor sizes with great precision. By combining Tuba-seq with multiplexed CRISPR/Cas9-mediated genome editing, we quantified the effects of eleven tumor-suppressor pathways that are frequently altered in human lung adenocarcinoma. With unprecedented resolution, parallelization, and precision Tuba-seq enables broad quantification of tumor suppressor gene function.