Peter Van Loo diagram.

Introduction

Our group aims to leverage the wealth of data from massively parallel sequencing efforts to understand carcinogenesis and cancer evolution.

Phylogenetic tree of a breast cancer, derived from massively parallel sequencing data.

Figure 1: Phylogenetic tree of a breast cancer, derived from massively parallel sequencing data.

Cancer is caused by somatic changes to the genome. Breakthrough sequencing technologies now enable us to compare the genomes of cancer cells to those of normal cells to base-pair resolution. This in theory allows us to identify all point mutations, small insertions and deletions, genomic rearrangements and copy number changes in a cancer cell, providing a unique opportunity to look at the genes and processes involved in cancer.

As genomic changes occur throughout a tumour's lifetime, a cancer's genome is also an archaeological record of its past, allowing a unique view of the tumour's evolutionary history.

Our group focuses on integrating cancer genomics data across tumour types to gain insight into the genes involved in cancer and into tumour evolution. This includes characterising the landscape of cancer genes, with a focus on identifying rare tumour suppressors through pan-cancer approaches integrating copy-number data with other mutation classes.

In addition, we are developing "molecular archaeology" approaches that use cancer genomes to obtain detailed timelines of cancer development across thousands of cancers.

As the cancer genome gradually reveals its secrets, we also focus on the next challenge: to understand how genomic changes lead to transcriptomic (and proteomic, interactomic, etc) changes to finally cause cancer. As a first step to tackle this, we are developing approaches to disentangle transcriptomes of admixed normal cells from cancer cell transcriptomes. We will next use these methods to study the relationship between genomic changes and the transcriptome of cancer cells.