Abstract

Breast cancer is a heterogeneous disease that has been extensively profiled by high throughput technologies and can be classified into different subtypes. However, the underlying causes of subtype differentiation remain unclear. Traditional experimental approaches to understand causality in breast cancer subtype differentiation are limited. This is due to the correlative nature of the data produced from profiling biological systems without perturbations, the use of low throughput single gene or gene product perturbation systems, and the time and resource bottleneck associated with traditional experimental approaches. To address these problems, we designed and implemented COIN-seq, a pooled COmbinatorial INtervention sequencing (COIN-seq) screen. COIN-seq combines single cell RNA sequencing and clustered regularly interspaced short palindromic repeats (CRISPR) based genetic interventions to perform massively parallel multi-locus gene perturbations. I contribute to the development of COIN-seq by aiding in the design and implementation of the different components involved with the system. In this thesis, we begin by describing the background information on breast cancer and the biological targets. We then explore the different technologies COIN-seq is predicated upon as well as the overall design and implementation of the system. Lastly, we discuss the successes and failures involved in this development process. Although a full implementation of this method was not achieved here, the successes and failures reported in this thesis can nevertheless serve as a guide for future development.