Search for: All records

ABSTRACT Haemophilus ducreyicauses the genital ulcer disease chancroid and cutaneous ulcers in children. To study its pathogenesis, we developed a human challenge model in which we infect the skin on the upper arm of human volunteers withH. ducreyito the pustular stage of disease. The model has been used to define lesional architecture, describe the immune infiltrate into the infected sites using flow cytometry, and explore the molecular basis of the immune response using bulk RNA-seq. Here, we used single cell RNA-seq (scRNA-seq) and spatial transcriptomics to simultaneously characterize multiple cell types within infected human skin and determine the cellular origin of differentially expressed transcripts that we had previously identified by bulk RNA-seq. We obtained paired biopsies of pustules and wounded (mock infected) sites from five volunteers for scRNA-seq. We identified 13 major cell types, including T- and NK-like cells, macrophages, dendritic cells, as well as other cell types typically found in the skin. Immune cell types were enriched in pustules, and some subtypes within the major cell types were exclusive to pustules. Sufficient tissue specimens for spatial transcriptomics were available from four of the volunteers. T- and NK-like cells were highly associated with multiple antigen presentation cell types. In pustules, type I interferon stimulation was high in areas that were high in antigen presentation—especially in macrophages near the abscess—compared to wounds. Together, our data provide a high-resolution view of the cellular immune response to the infection of the skin with a human pathogen.IMPORTANCEA high-resolution view of the immune infiltrate due to infection with an extracellular bacterial pathogen in human skin has not yet been defined. Here, we used the human skin pathogenHaemophilus ducreyiin a human challenge model to identify on a single cell level the types of cells that are present in volunteers who fail to spontaneously clear infection and form pustules. We identified 13 major cell types. Immune cells and immune-activated stromal cells were enriched in pustules compared to wounded (mock infected) sites. Pustules formed despite the expression of multiple pro-inflammatory cytokines, such as IL-1β and type I interferon. Interferon stimulation was most evident in macrophages, which were proximal to the abscess. The pro-inflammatory response within the pustule may be tempered by regulatory T cells and cells that express indoleamine 2,3-dioxygenase, leading to failure of the immune system to clearH. ducreyi. 

Matrix low rank approximation is an effective method to reduce or eliminate the statistical redundancy of its components. Compared with the traditional global low rank methods such as singular value decomposition (SVD), local low rank approximation methods are more advantageous to uncover interpretable data structures when clear duality exists between the rows and columns of the matrix. Local low rank approximation is equivalent to low rank submatrix detection. Unfortunately,existing local low rank approximation methods can detect only submatrices of specific mean structure, which may miss a substantial amount of true and interesting patterns. In this work, we develop a novel matrix computational framework called RPSP (Random Probing based submatrix Propagation) that provides an effective solution for the general matrix local low rank representation problem. RPSP detects local low rank patterns that grow from small submatrices of low rank property, which are determined by a random projection approach. RPSP is supported by theories of random projection. Experiments on synthetic data demonstrate that RPSP outperforms all state-of-the-art methods, with the capacity to robustly and correctly identify the low rank matrices when the pattern has a similar mean as the background, background noise is heteroscedastic and multiple patterns present in the data. On real-world datasets, RPSP also demonstrates its effectiveness in identifying interpretable local low rank matrices.