Abstract Acute Myeloid Leukaemia (AML) is a heterogeneous cancer of the blood affecting the normal development of white blood cells and with patient’s prognosis varying dramatically between different subtypes. AML patients harbour a low number of mutations compared to other cancers while the expression of thousands of genes is often dysregulated. This makes RNA sequencing (RNA-Seq) an appealing instrument to track both genetic and transcriptional determinants of response in leukaemia.
During my PhD, I tackled relapse in AML by studying two cohorts of bulk RNA-Seq samples, collected as part of two Australian clinical trials and with samples available over time. I developed a bioinformatic workflow and applied statistical methods to extract and analyse both gene expression and somatic mutations from RNA-Seq, which required developing new analytical approaches. The workflow was applied to a published AML dataset to help guide both the experimental design and the analysis of both Australian cohorts. My work also tackled confounding factors in the gene expression data due to the varying mixture of tumour and normal cells as well as strategies to efficiently visualise the results to gain insight into factors that impact on therapy response and patient outcome.