Data availability. All of the sequencing data is available via Gene Expression Omnibus (GEO) under the accession number GSE117826.

Single-cell sequencing, which is used to detect clinically important tumor subpopulations, is necessary for understanding tumor heterogeneity. Here, we analyzed transcriptomic data obtained from 34 single cells from human lung adenocarcinoma (LADC) patient-derived xenografts (PDXs). To focus on the intrinsic transcriptomic signatures of these tumors, we filtered out genes that displayed extensive expression changes following xenografting and cell culture. Then, we performed clustering analysis using co-regulated gene modules rather than individual genes to minimize read drop-out errors associated with single-cell sequencing. This combined approach revealed two distinct intra-tumoral subgroups that were primarily distinguished by the gene module G64. The G64 module was predominantly composed of cell-cycle genes. E2F1 was found to be the transcription factor that most likely mediates the expression of the G64 module in single LADC cells. Interestingly, the G64 module also indicated inter-tumoral heterogeneity based on its association with patient survival and other clinical variables such as smoking status and tumor stage. Taken together, these results demonstrate the feasibility of single-cell RNA sequencing and the strength of our analytical pipeline for the identification of tumor subpopulations.

Defining the transcriptomic identity of clonally related malignant cells is challenging in the absence of cell surface markers that distinguish cancer clones from one another or from admixed non-neoplastic cells. While single-cell methods have been devised to capture both the transcriptome and genotype, these methods are not compatible with droplet-based single-cell transcriptomics, limiting their throughput. To overcome this limitation, we present single-cell Genotyping of Transcriptomes (GoT), which integrates cDNA genotyping with high-throughput droplet-based single-cell RNA-seq. We further demonstrate that multiplexed GoT can interrogate multiple genotypes for distinguishing subclonal transcriptomic identity. We apply GoT to 26,039 CD34 + cells across six patients with myeloid neoplasms, in which the complex process of hematopoiesis is corrupted by CALR-mutated stem and progenitor cells. We define highresolution maps of malignant versus normal hematopoietic progenitors, and show that while mutant cells are comingled with wildtype cells throughout the hematopoietic progenitor landscape, their frequency increases with differentiation. We identify the unfolded protein response as a predominant outcome of CALR mutations, with significant cell identity dependency. Furthermore, we identify that CALR mutations lead to NF-kB pathway upregulation specifically in uncommitted early stem cells. Collectively, GoT provides highthroughput linkage of single-cell genotypes with transcriptomes and reveals that the transcriptional output of somatic mutations is heavily dependent on the native cell identity.Somatic mutations underlie the development of clonal outgrowth and malignant transformation 1-5 . Ongoing clonal evolution through acquisition of further genetic alterations often result in multiclonal cancer populations 6-10 . Transcriptional read-outs are critical for the study of the molecular basis of these processes. However, clonally-derived populations often lack cell surface markers that distinguish them from normal cells or that can help distinguish subclones, limiting the effectiveness of bulk RNA sequencing for investigating the clonal architecture of malignant cell populations. For example, myeloproliferative neoplasms (MPNs) have been shown to result from recurrent somatic mutations in JAK2, CALR and MPL, which are present across multiple progenitor classes, including CD34 + , CD38hematopoietic stem progenitor cells (HSPC) and downstream progenitor cells including megakaryocytic-erythroid progenitors (MEP) 11,12 . These malignant clones often represent a subset of the bone marrow progenitor population without distinctive cell surface markers to distinguish them from non-neoplastic hematopoietic cells.Furthermore, MPNs are known to undergo clonal diversification with subclones harboring driver gene mutations 11,13,14 , without a tractable avenue for isolation of subclones based on cell surface markers. Therefore, our current understanding of the impact of the somatic mutations in MPN has relied primarily on cell l...