Approaches for identifying lung cell type responses to perturbation

Abstract

The use of genomic profiling can provide indications of underlying molecular responses to chemical perturbation, and the characterization of these responses can provide an increased understanding of the greater physiological effects of an exposure and inform clinical decision making. This approach has proven to be effective in understanding the effects of environmental exposures such as cigarette smoke on the airway epithelium, and how they may contribute to associated disease pathogenesis. Because of the existing body of work in genomic profiling towards understanding the effects of environmental exposures, it has relevant applications towards the study of the effects of emerging exposures such as electronic cigarettes, which remain poorly understood. Further, current approaches for genomic profiling could be improved through the development of data resources and computational methods which can identify not only tissue- or sample-level molecular responses to perturbation, but also responses specific to individual cells or cell types.

In light of these issues, I investigated the molecular response in airway epithelium to a novel inhaled exposure, and developed methods to support more detailed characterization of such effects. In this dissertation, I describe a clinical observational study in which I examined the gene expression effects of electronic cigarettes on the airway epithelium, and compare these effects to those of conventional cigarettes (Aim 1). Next, I describe CELDA, a novel computational method for identifying cell subpopulations and the co-expressed modules of genes that identify them in single cell RNA-seq (scRNA-seq) data (Aim 2). Finally, I describe the Lung Connectivity Map (Lung CMap), a platform for interrogating lung cell type specific responses to a large set of chemical and molecular perturbations (Aim 3).

Collectively, this work encompasses both observational and computational approaches for detailed characterization of the molecular responses to perturbation, and the determination of the relative effects of these novel perturbations versus their more well-described counterparts.