Current inflammatory bowel disease (IBD) therapies are ineffective in a high proportion of patients. Combining bulk and single-cell transcriptomics, quantitative histopathology and in situ localization across three cohorts of patients with IBD (total n = 376), we identify coexpressed gene modules within the heterogeneous tissular inflammatory response in IBD that map to distinct histopathological and cellular features (pathotypes). One of these pathotypes is defined by high neutrophil infiltration, activation of fibroblasts and vascular remodeling at sites of deep ulceration. Activated fibroblasts in the ulcer bed display neutrophil-chemoattractant properties that are IL-1R, but not TNF, dependent. Pathotype-associated neutrophil and fibroblast signatures are increased in nonresponders to several therapies across four independent cohorts (total n = 343). The identification of distinct, localized, tissular pathotypes will aid precision targeting of current therapeutics and provides a biological rationale for IL-1 signaling blockade in ulcerating disease.
Recent advances in single-cell technologies and integration algorithms make it possible to construct comprehensive reference atlases encompassing many donors, studies, disease states, and sequencing platforms. Much like mapping sequencing reads to a reference genome, it is essential to be able to map query cells onto complex, multimillion-cell reference atlases to rapidly identify relevant cell states and phenotypes. We present Symphony (https://github.com/immunogenomics/symphony), an algorithm for building large-scale, integrated reference atlases in a convenient, portable format that enables efficient query mapping within seconds. Symphony localizes query cells within a stable low-dimensional reference embedding, facilitating reproducible downstream transfer of reference-defined annotations to the query. We demonstrate the power of Symphony in multiple real-world datasets, including (1) mapping a multi-donor, multi-species query to predict pancreatic cell types, (2) localizing query cells along a developmental trajectory of fetal liver hematopoiesis, and (3) inferring surface protein expression with a multimodal CITE-seq atlas of memory T cells.
How innate T cells (ITC), including invariant natural killer T (iNKT) cells, mucosal-associated invariant T (MAIT) cells, and γδ T cells, maintain a poised effector state has been unclear. Here we address this question using low-input and single-cell RNA-seq of human lymphocyte populations. Unbiased transcriptomic analyses uncover a continuous ‘innateness gradient’, with adaptive T cells at one end, followed by MAIT, iNKT, γδ T and natural killer cells at the other end. Single-cell RNA-seq reveals four broad states of innateness, and heterogeneity within canonical innate and adaptive populations. Transcriptional and functional data show that innateness is characterized by pre-formed mRNA encoding effector functions, but impaired proliferation marked by decreased baseline expression of ribosomal genes. Together, our data shed new light on the poised state of ITC, in which innateness is defined by a transcriptionally-orchestrated trade-off between rapid cell growth and rapid effector function.
Summary: The related Wilcoxon rank sum test and area under the receiver operator curve are ubiquitous in high dimensional biological data analysis. Current implementations do not scale readily to the increasingly large datasets generated by novel high-throughput technologies, such as single cell RNAseq. We introduce a simple and scalable implementation of both analyses, available through the R package Presto. Presto scales to big datasets, with functions optimized for both dense and sparse matrices. On a sparse dataset of 1 million observations, 10 groups, and 1,000 features, Presto performed both rank-sum and auROC analyses in only 17 seconds, compared to 6.4 hours with base R functions. Presto also includes functions to seamlessly integrate with the Seurat single cell analysis pipeline and the Bioconductor SingleCellExperiment class. Presto enables the use of robust classical analyses on big data with a simple interface and optimized implementation. Availability and Implementation: Presto is available as an R package at https://github.com/immunogenomics/presto . Contact: soumya@broadinstitute.org Supplementary Information: Vignettes are available with the Presto package.