Expansion and remodelling of the mammary epithelium requires a tight balance between cellular proliferation, differentiation and death. To explore cell survival versus cell death decisions in this organ, we deleted the pro-survival gene Mcl-1 in the mammary epithelium. Mcl-1 was found to be essential at multiple developmental stages including morphogenesis in puberty and alveologenesis in pregnancy. Moreover, Mcl-1-deficient basal cells were virtually devoid of repopulating activity, suggesting that this gene is required for stem cell function. Profound upregulation of the Mcl-1 protein was evident in alveolar cells at the switch to lactation, and Mcl-1 deficiency impaired lactation. Interestingly, EGF was identified as one of the most highly upregulated genes on lactogenesis and inhibition of EGF or mTOR signalling markedly impaired lactation, with concomitant decreases in Mcl-1 and phosphorylated ribosomal protein S6. These data demonstrate that Mcl-1 is essential for mammopoiesis and identify EGF as a critical trigger of Mcl-1 translation to ensure survival of milk-producing alveolar cells.

Single-cell RNA sequencing (scRNA-seq) is widely used to profile the transcriptome of individual cells. This provides biological resolution that cannot be matched by bulk RNA sequencing, at the cost of increased technical noise and data complexity. The differences between scRNA-seq and bulk RNA-seq data mean that the analysis of the former cannot be performed by recycling bioinformatics pipelines for the latter. Rather, dedicated single-cell methods are required at various steps to exploit the cellular resolution while accounting for technical noise. This article describes a computational workflow for low-level analyses of scRNA-seq data, based primarily on software packages from the open-source Bioconductor project. It covers basic steps including quality control, data exploration and normalization, as well as more complex procedures such as cell cycle phase assignment, identification of highly variable and correlated genes, clustering into subpopulations and marker gene detection. Analyses were demonstrated on gene-level count data from several publicly available datasets involving haematopoietic stem cells, brain-derived cells, T-helper cells and mouse embryonic stem cells. This will provide a range of usage scenarios from which readers can construct their own analysis pipelines.

Normalization of single-cell RNA sequencing data is necessary to eliminate cell-specific biases prior to downstream analyses. However, this is not straightforward for noisy single-cell data where many counts are zero. We present a novel approach where expression values are summed across pools of cells, and the summed values are used for normalization. Pool-based size factors are then deconvolved to yield cell-based factors. Our deconvolution approach outperforms existing methods for accurate normalization of cell-specific biases in simulated data. Similar behavior is observed in real data, where deconvolution improves the relevance of results of downstream analyses.

Barcode swapping results in the mislabelling of sequencing reads between multiplexed samples on patterned flow-cell Illumina sequencing machines. This may compromise the validity of numerous genomic assays; however, the severity and consequences of barcode swapping remain poorly understood. We have used two statistical approaches to robustly quantify the fraction of swapped reads in two plate-based single-cell RNA-sequencing datasets. We found that approximately 2.5% of reads were mislabelled between samples on the HiSeq 4000, which is lower than previous reports. We observed no correlation between the swapped fraction of reads and the concentration of free barcode across plates. Furthermore, we have demonstrated that barcode swapping may generate complex but artefactual cell libraries in droplet-based single-cell RNA-sequencing studies. To eliminate these artefacts, we have developed an algorithm to exclude individual molecules that have swapped between samples in 10x Genomics experiments, allowing the continued use of cutting-edge sequencing machines for these assays.

Loss of cone photoreceptors, crucial for daylight vision, has the greatest impact on sight in retinal degeneration. Transplantation of stem cell-derived L/M-opsin cones, which form 90% of the human cone population, could provide a feasible therapy to restore vision. However, transcriptomic similarities between fetal and stem cell-derived cones remain to be defined, in addition to development of cone cell purification strategies. Here, we report an analysis of the human L/M-opsin cone photoreceptor transcriptome using an AAV2/9.pR2.1:GFP reporter. This led to the identification of a cone-enriched gene signature, which we used to demonstrate similar gene expression between fetal and stem cell-derived cones. We then defined a cluster of differentiation marker combination that, when used for cell sorting, significantly enriches for cone photoreceptors from the fetal retina and stem cell-derived retinal organoids, respectively. These data may facilitate more efficient isolation of human stem cell-derived cones for use in clinical transplantation studies.

By profiling the transcriptomes of individual cells, single-cell RNA sequencing provides unparalleled resolution to study cellular heterogeneity. However, this comes at the cost of high technical noise, including cell-specific biases in capture efficiency and library generation. One strategy for removing these biases is to add a constant amount of spike-in RNA to each cell and to scale the observed expression values so that the coverage of spike-in transcripts is constant across cells. This approach has previously been criticized as its accuracy depends on the precise addition of spike-in RNA to each sample. Here, we perform mixture experiments using two different sets of spike-in RNA to quantify the variance in the amount of spike-in RNA added to each well in a plate-based protocol. We also obtain an upper bound on the variance due to differences in behavior between the two spike-in sets. We demonstrate that both factors are small contributors to the total technical variance and have only minor effects on downstream analyses, such as detection of highly variable genes and clustering. Our results suggest that scaling normalization using spike-in transcripts is reliable enough for routine use in single-cell RNA sequencing data analyses.

How cells respond to myriad stimuli with finite signaling machinery is central to immunology. In naive T cells, the inherent effect of ligand strength on activation pathways and endpoints has remained controversial, confounded by environmental fluctuations and intercellular variability within populations. Here we studied how ligand potency affected the activation of CD8+ T cells in vitro, through the use of genome-wide RNA, multi-dimensional protein and functional measurements in single cells. Our data revealed that strong ligands drove more efficient and uniform activation than did weak ligands, but all activated cells were fully cytolytic. Notably, activation followed the same transcriptional pathways regardless of ligand potency. Thus, stimulation strength did not intrinsically dictate the T cell-activation route or phenotype; instead, it controlled how rapidly and simultaneously the cells initiated activation, allowing limited machinery to elicit wide-ranging responses.

The structural flexibility of RNA underlies fundamental biological processes, but there are no methods for exploring the multiple conformations adopted by RNAs in vivo. We developed cross-linking of matched RNAs and deep sequencing (COMRADES) for in-depth RNA conformation capture, and a pipeline for the retrieval of RNA structural ensembles. Using COMRADES, we determined the architecture of the Zika virus RNA genome inside cells, and identified multiple site-specific interactions with human noncoding RNAs.

An increasing number of studies are using single-cell RNA-sequencing (scRNA-seq) to characterize the gene expression profiles of individual cells. One common analysis applied to scRNA-seq data involves detecting differentially expressed (DE) genes between cells in different biological groups. However, many experiments are designed such that the cells to be compared are processed in separate plates or chips, meaning that the groupings are confounded with systematic plate effects. This confounding aspect is frequently ignored in DE analyses of scRNA-seq data. In this article, we demonstrate that failing to consider plate effects in the statistical model results in loss of type I error control. A solution is proposed whereby counts are summed from all cells in each plate and the count sums for all plates are used in the DE analysis. This restores type I error control in the presence of plate effects without compromising detection power in simulated data. Summation is also robust to varying numbers and library sizes of cells on each plate. Similar results are observed in DE analyses of real data where the use of count sums instead of single-cell counts improves specificity and the ranking of relevant genes. This suggests that summation can assist in maintaining statistical rigour in DE analyses of scRNA-seq data with plate effects.

Data exploration is critical to the comprehension of large biological data sets generated by high-throughput assays such as sequencing. However, most existing tools for interactive visualisation are limited to specific assays or analyses. Here, we present the iSEE (Interactive SummarizedExperiment Explorer) software package, which provides a general visual interface for exploring data in a SummarizedExperiment object. iSEE is directly compatible with many existing R/Bioconductor packages for analysing high-throughput biological data, and provides useful features such as simultaneous examination of (meta)data and analysis results, dynamic linking between plots and code tracking for reproducibility. We demonstrate the utility and flexibility of iSEE by applying it to explore a range of real transcriptomics and proteomics data sets.

Blood-borne pathogens can cause systemic inflammatory response syndrome (SIRS) followed by protracted, potentially lethal immunosuppression. The mechanisms responsible for impaired immunity post-SIRS remain unclear. We show that SIRS triggered by pathogen mimics or malaria infection leads to functional paralysis of conventional dendritic cells (cDCs). Paralysis affects several generations of cDCs and impairs immunity for 3-4 weeks. Paralyzed cDCs display distinct transcriptomic and phenotypic signatures and show impaired capacity to capture and present antigens in vivo. They also display altered cytokine production patterns upon stimulation. The paralysis program is not initiated in the bone marrow but during final cDC differentiation in peripheral tissues under the influence of local secondary signals that persist after resolution of SIRS. Vaccination with monoclonal antibodies that target cDC receptors or blockade of transforming growth factor β partially overcomes paralysis and immunosuppression. This work provides insights into the mechanisms of paralysis and describes strategies to restore immunocompetence post-SIRS.

Epigenetic mechanisms involved in the establishment of lung epithelial cell lineage identities during development are largely unknown. Here, we explored the role of the histone methyltransferase Ezh2 during lung lineage determination. Loss of Ezh2 in the lung epithelium leads to defective lung formation and perinatal mortality. We show that Ezh2 is crucial for airway lineage specification and alveolarization. Using optical projection tomography imaging, we found that branching morphogenesis is affected in Ezh2 conditional knockout mice and the remaining bronchioles are abnormal, lacking terminally differentiated secretory club cells. Remarkably, RNA-seq analysis revealed the upregulation of basal genes in Ezh2-deficient epithelium. Three-dimensional imaging for keratin 5 further showed the unexpected presence of a layer of basal cells from the proximal airways to the distal bronchioles in E16.5 embryos. ChIP-seq analysis indicated the presence of Ezh2-mediated repressive marks on the genomic loci of some but not all basal genes, suggesting an indirect mechanism of action of Ezh2. We found that loss of Ezh2 de-represses insulin-like growth factor 1 (Igf1) expression and that modulation of IGF1 signaling ex vivo in wild-type lungs could induce basal cell differentiation. Altogether, our work reveals an unexpected role for Ezh2 in controlling basal cell fate determination in the embryonic lung endoderm, mediated in part by repression of Igf1 expression.

In recent years, RNA sequencing (RNA-seq) has become a very widely used technology for profiling gene expression. One of the most common aims of RNA-seq profiling is to identify genes or molecular pathways that are differentially expressed (DE) between two or more biological conditions. This article demonstrates a computational workflow for the detection of DE genes and pathways from RNA-seq data by providing a complete analysis of an RNA-seq experiment profiling epithelial cell subsets in the mouse mammary gland. The workflow uses R software packages from the open-source Bioconductor project and covers all steps of the analysis pipeline, including alignment of read sequences, data exploration, differential expression analysis, visualization and pathway analysis. Read alignment and count quantification is conducted using the Rsubread package and the statistical analyses are performed using the edgeR package. The differential expression analysis uses the quasi-likelihood functionality of edgeR.

Chromatin immunoprecipitation with massively parallel sequencing (ChIP-seq) is widely used to identify binding sites for a target protein in the genome. An important scientific application is to identify changes in protein binding between different treatment conditions, i.e. to detect differential binding. This can reveal potential mechanisms through which changes in binding may contribute to the treatment effect. The csaw package provides a framework for the de novo detection of differentially bound genomic regions. It uses a window-based strategy to summarize read counts across the genome. It exploits existing statistical software to test for significant differences in each window. Finally, it clusters windows into regions for output and controls the false discovery rate properly over all detected regions. The csaw package can handle arbitrarily complex experimental designs involving biological replicates. It can be applied to both transcription factor and histone mark datasets, and, more generally, to any type of sequencing data measuring genomic coverage. csaw performs favorably against existing methods for de novo DB analyses on both simulated and real data. csaw is implemented as a R software package and is freely available from the open-source Bioconductor project.

When comparing biological conditions using mass cytometry data, a key challenge is to identify cellular populations that change in abundance. Here, we present a computational strategy for detecting 'differentially abundant' populations by assigning cells to hyperspheres, testing for significant differences between conditions and controlling the spatial false discovery rate. Our method (http://bioconductor.org/packages/cydar) outperforms other approaches in simulations and finds novel patterns of differential abundance in real data.

Single-cell RNA sequencing (scRNA-seq) is increasingly used to study gene expression at the level of individual cells. However, preparing raw sequence data for further analysis is not a straightforward process. Biases, artifacts and other sources of unwanted variation are present in the data, requiring substantial time and effort to be spent on pre-processing, quality control (QC) and normalization.

It has been proposed that interactions between mammalian chromosomes, or transchromosomal interactions (also known as kissing chromosomes), regulate gene expression and cell fate determination. Here we aimed to identify novel transchromosomal interactions in immune cells by high-resolution genome-wide chromosome conformation capture. Although we readily identified stable interactions in cis, and also between centromeres and telomeres on different chromosomes, surprisingly we identified no gene regulatory transchromosomal interactions in either mouse or human cells, including previously described interactions. We suggest that advances in the chromosome conformation capture technique and the unbiased nature of this approach allow more reliable capture of interactions between chromosomes than previous methods. Overall our findings suggest that stable transchromosomal interactions that regulate gene expression are not present in mammalian immune cells and that lineage identity is governed by cis, not trans chromosomal interactions.

Chromatin immunoprecipitation with massively parallel sequencing (ChIP-seq) is widely used to identify the genomic binding sites for protein of interest. Most conventional approaches to ChIP-seq data analysis involve the detection of the absolute presence (or absence) of a binding site. However, an alternative strategy is to identify changes in the binding intensity between two biological conditions, i.e., differential binding (DB). This may yield more relevant results than conventional analyses, as changes in binding can be associated with the biological difference being investigated. The aim of this article is to facilitate the implementation of DB analyses, by comprehensively describing a computational workflow for the detection of DB regions from ChIP-seq data. The workflow is based primarily on R software packages from the open-source Bioconductor project and covers all steps of the analysis pipeline, from alignment of read sequences to interpretation and visualization of putative DB regions. In particular, detection of DB regions will be conducted using the counts for sliding windows from the csaw package, with statistical modelling performed using methods in the edgeR package. Analyses will be demonstrated on real histone mark and transcription factor data sets. This will provide readers with practical usage examples that can be applied in their own studies.

Droplet-based single-cell RNA sequencing protocols have dramatically increased the throughput of single-cell transcriptomics studies. A key computational challenge when processing these data is to distinguish libraries for real cells from empty droplets. Here, we describe a new statistical method for calling cells from droplet-based data, based on detecting significant deviations from the expression profile of the ambient solution. Using simulations, we demonstrate that EmptyDrops has greater power than existing approaches while controlling the false discovery rate among detected cells. Our method also retains distinct cell types that would have been discarded by existing methods in several real data sets.

Transposable elements (TEs) regulate diverse biological processes, from early development to cancer. Expression of young TEs is difficult to measure with next-generation, single-cell sequencing technologies because their highly repetitive nature means that short complementary DNA reads cannot be unambiguously mapped to a specific locus. Single CELl LOng-read RNA-sequencing (CELLO-seq) combines long-read single cell RNA-sequencing with computational analyses to measure TE expression at unique loci. We used CELLO-seq to assess the widespread expression of TEs in two-cell mouse blastomeres as well as in human induced pluripotent stem cells. Across both species, old and young TEs showed evidence of locus-specific expression with simulations demonstrating that only a small number of very young elements in the mouse could not be mapped back to the reference with high confidence. Exploring the relationship between the expression of individual elements and putative regulators revealed large heterogeneity, with TEs within a class showing different patterns of correlation and suggesting distinct regulatory mechanisms.